Printer Friendly

Leaf and root production, decomposition and carbon and nitrogen fluxes during stand development in tropical moist forests, north-east India.

Abstract. In the present study we investigated the production and decomposition of leaves, branches and roots in two moist regenerating tropical forests (5 and 15 years old; RF-5 and RF-15 respectively) and a natural forest (NF) in north-east India. Total litter input increased during vegetation succession (759, 1089 and 1284 g [m.sup.-1] [year.sup.-1] in RF-5, RF-15 and NF respectively), whereas the contribution of soft litter decreased sharply. Decomposition over 450 days indicated significant seasonal (P < 0.001) patterns in mass loss of litter components, with greater rates during the wet period. Soil C[O.sub.2] efflux was strongly seasonal. C stock loss followed patterns similar to those of mass loss, whereas N increased initially, followed by its gradual release. Rainfall explained 74-90% of the variability in mass loss rates. Concentrations of cellulose and N were significantly positively correlated with mass loss at an early stage of decomposition (r=0.54-0.65, P<0.05), whereas lignin: N and C : N ratios were negatively correlated with mass loss at later stages. Regenerating forests adapted ecosystem-level strategies that induced early leaf fall to reduce soil water loss, increase organic matter return to the soil and conserve N through immobilisation during the process of decomposition to speed up vegetation succession in the regenerating forest.

Additional keywords: carbon and nitrogen dynamics, litter production, natural tropical forest, regenerating forests.

Introduction

Land use and land cover changes are among the most important noticeable changes occurring around us as a result of anthropogenic and environmental factors, and quantitative assessment of these changes has become a challenging task because of their magnitude, variety and spatial variability (Roy and Roy 2010). Tropical forests are among the most important biomes because they store over half the Earth's biodiversity (Lewis et al. 2015), maintain critical ecosystem services, exchange large quantities of carbon (C) and water (Slik et al. 2013; Lewis et al. 2015) and cycle nutrients (Kotowska et al. 2015). Forests in tropical regions are threatened by widespread deforestation due to increasing human population and demands for agricultural products (Lewis et al. 2015). This has led to rapid land use change from natural forest to a multitude of ecosystem types, resulting in dominant landscape types under secondary forests following disturbance (Aide and Grau 2004; Chazdon 2014). These secondary forests can provide many ecosystem services, including the potential to function as sinks for atmospheric C in plant biomass and soil (Silver et al. 2004). Patterns and dynamics of natural succession differ greatly at the stand scale within and between regions (van Breugel et al. 2007), and the recovery of tropical forests can be strongly limited by the range of biotic and abiotic factors, particularly decreased soil nutrient availability (Holl 2002). Therefore, forest restoration approaches may vary depending on levels of site degradation, residual vegetation, desired outcomes and budget (Chazdon et al. 2008; Celentano et al. 2011). Although structural changes in species composition and biomass accumulation during the course of ecosystem development are well studied (Denslow and Guzman 2000; Martin et al. 2004; Singh et al. 2015), there are only few investigations regarding the functioning of secondary forests (Feldpausch et al. 2004; Chazdon 2014), and no studies are available in Mizoram, northeast India. Therefore, the aim of the present study was to provide baseline data for use in developing policies and models that can be used for restoration programs of tropical forests.

Most of tropical forests occur in soils with moderate to low fertility and significant quantities of nutrients are stored in the plant biomass, which is potentially available to the biota, and plants have developed adaptive mechanisms for the acquisition and retention of nutrients (Lambers et al. 2008). Consequently, tropical areas are vulnerable to deforestation due to disrupted nutrient cycling and increased nutrient leakage from the system, in addition to the removal of nutrient capital (Walker and Reddel 2007). Despite the significant role played by secondary forests in C and nutrient cycling, there are only a few studies on litterfall and decomposition (Xuluc-Tolosa et al. 2003; Ostertag et al. 2008; Celentano et al. 2011). Our understanding of the controls on plant C inputs to the forest soil and the rates of decomposition of litter material during secondary succession is limited (Ostertag et al. 2008).

Litter production and decomposition represents one of the most important pathways for the transfer of energy and material in forest ecosystems (Pandey et al. 2007; Gavazov et al. 2014). Thus, litter input and decomposition are important processes responsible for regulating productivity and nutrient cycling, and help in forest recovery (Loaiza-Usuga et al. 2013; Pandey et al. 2016). The belowground litters act as an important source of C and N and serves as a key mechanism for the restoration of degraded tropical forests (Pandey et al. 2016). In the tropics, seasonal patterns of litter production and decomposition have been widely reported to regulate soil fertility and the rate of nutrient availability (Tripathi and Singh 1992a; Fioretto et al. 2003; Chen et al. 2014; Gavazov et al. 2014). Forest degradation alters vegetation composition, litter input and decomposition rates, which affect forest microclimate and litter chemistry (Mtambanengwe and Kirchman 1995) and finally soil fertility (Fioretto et al. 2003). Recently, C and nutrient cycling has been reported to be more strongly affected by climate seasonality in species-poor compared with species-rich ecosystems (Kotowska et al. 2015). Therefore, land use change may have more pronounced effects on the cycling of C and nutrients in degraded tropical ecosystems.

Litter decomposition is governed by groups of factors (e.g. climate, initial litter quality and micro-organisms; Tripathi and Singh 19926; Gavazov etal. 2014). Among the climatic factors, precipitation and its associated variables (i.e. soil moisture, humidity) and temperature were found to significantly affect litter decomposition in different forest ecosystems by providing moisture and maintaining a favourable microclimate that promotes microbial activity (Boyero et al. 2014; Purahong et al. 2015). The initial litter quality variables (labile fractions, cellulose, lignin, C and nitrogen (N) content and C: N and lignin: N ratios) play a crucial role at different stages of litter decomposition by providing available sources of energy for the decomposers in many ecosystems (Zhang et al. 2013). There are abundant studies on the decomposition of aboveground litter components in temperate ecosystems, but fewer studies in tropical ecosystems. Furthermore, studies on belowground litter decomposition, particularly root litter fluxes, have often been ignored in many ecosystems despite its significant role in regulating C and N cycling.

Mizoram, one of the seven states of north-east India, covering an area of 21 081 [km.sup.2], exhibits three vegetation types: tropical wet evergreen forest, tropical semi-evergreen forest and subtropical pine forest (Champion and Seth 1968). The recorded forest area of the state is 90% of its total geographical area (State Statistical Handbook of Mizoram 2014). The rich wealth of the tropical forests experienced degradation due to overexploitation of forest land for timber, firewood, construction, mining and the practice of shifting cultivation. As a result of overexploitation of forest, the area under degraded forest has increased and is at different stages of secondary succession. Studies have shown that species composition and their relative abundance change during the course of succession; for example, young secondary successional forest contains more herbaceous components, with an increased proportion of woody species in the later stages of forest development (Singh et al. 2015). Given these changes in species composition, we expect changes in forest microclimate, litter input, chemistry and decomposition during secondary succession. We hypothesised that forest disturbance will reduce litter production (leaf and root), decomposition and C and N stocks during the early stages of stand development. To test this hypothesis, in the present study we measured leaf litter and root production, decomposition, litter chemistry and soil C[O.sub.2] efflux during ecosystem development following stone mining in Mizoram, north-east India.

Materials and methods

Study sites

The present experiment was conducted in three sites. A 1-ha area was selected in each site from a large tract. At each site, five permanent plots (20 m x 20 m) were randomly demarcated for intensive periodic sampling of roots, litter and soil. Two sites were 5- and 15-year-old regenerating forests (RF-5 and RF-15 respectively; 23[degrees]37'18"-23[degrees]39/55"'N, 92031'48"-92033'24"'E and 23[degrees]43'55"-23[degrees]44'48"N, 92[degrees]38'43"-92[degrees]40'4"E), where stone mining activity had taken place 5 and 15 years previously (i.e. in 2006 and 1996 respectively). Vegetation had developed naturally at these sites since the disturbance. The distance between the two regenerating sites is nearly 3 km. In addition, a nearby natural forest (NF) without disturbance in Tanhril (23[degrees]43'55"9'-23[degrees]44'48"N, 92[degrees]38'42"-92[degrees]40'6"E; 604-860 m above sea level), approximately 2 km from the disturbed sites, was selected. The area is ~15 km from Aizawl city and spreads across approximately 700 ha land. The region was previously supported by lush green vegetation dominated by tropical semi-evergreen forest (Champion and Seth 1968).

The three study sites have similar climatic conditions (i.e. moist subtropical) with regard to major climatic and edaphic factors, but differ in microclimatic conditions due to disturbances. Total annual rainfall ranged from 2100 to 2417 mm, of which approximately 80% occurred in 5 months (from May to September; Fig. 1). Relative humidity varied from 62% to 94% during the study period. The year has four distinct seasons: a cold and dry winter (mean temperature 16[degrees]C; December-February), a warm pre-monsoon (summer) period (mean temperature 22[degrees]C; March-May), a warm humid monsoon (mean temperature 25[degrees]C; June-September) and a cool post-monsoon (autumn) period (mean temperature 21[degrees]C; October-November).

The vegetation of the three forest sites was tropical semievergreen with distinct canopy cover (Table 1). The common species present at these sites were Schima wallichii, Sterculia villosa, Callicarpa arborea, Emblica officinalis, Albizzia chinensis, Castanopsis tribuloides, Rhus succedanea, Toona ciliata, Wendlandia tinctoria and Wendlandiagrandis (Singh et al. 2015). Grass species, such as Imperata cylindrica, Saccharum longisetosum and Thysanolaena maxima, were most abundant in recovering stands.

Soil characteristics and C[O.sub.2] efflux

In 2012-13, seasonal soil samples were collected from a depth of 0-10 cm four times a year (i.e.in January, April, July and October) from each site using a soil auger (diameter 5 cm). At each time point, three random soil samples (at least 5 m apart) were collected from each of the permanent 20-m x 20-m plots and pooled into a single sample. This way, 15 random soil samples were collected and combined into five samples, one each from five permanent plots. Soil samples were analysed seasonally in triplicate for physicochemical and biochemical properties. Soil pH was measured in a soil-water suspension (1:2.5 w/v [H.sub.2]O) using a digital pH meter. Soil moisture content was determined gravimetrically by oven drying the soil samples at 105[degrees]C as described by Anderson and Ingram (1993). Soil bulk density (g [cm.sup.-3]) was measured using a metallic tube of known inner volume and by estimating the dry weight of a unit volume of soil. Total N and C were analysed using the CHNS/O-Elemental Analyzer with autosampler (Euro Vector, Model: EuroEA3000, Mandaluyong City, Philippines) at the Central Instrumental Laboratory in Mizoram University. Available phosphorus ([P.sub.avail]) in the soil was determined using the stannous chloride blue colour method with a spectrophotometer (Page et al. 1982). Exchangeable potassium ([K.sub.exch]) was estimated using a flame photometer and soil extract of 1M ammonium acetate. Microbial biomass carbon (MBC) was measured by chloroform-fumigation and extraction as described previously (Vance et al. 1987), using the equation MBC = [E.sub.C]/[k.sub.EC], where EC is calculated by subtracting organic C extracted from non-fumigated soils from organic C extracted from fumigated soils, and [k.sub.EC] is the conversion factor for estimating carbon in microbial body as microbial body contains about 45% C in their dry mass (0.45).

Soil C[O.sub.2] efflux was measured from each site using the static alkali absorption method. At each time point, two random soil collars ~5m apart were inserted in each permanent 20-m x 20-m plot, with samples later combined into a single sample: 30 collars were fitted in 15 permanent plots at the three sites. Of the 30 collars, five replicated measurements of C[O.sub.2] efflux were recorded from each of the five permanent plots (during June 2012-August 2013) corresponding to the litter bag retrieval dates. For measurement of soil C[O.sub.2] efflux, polyvinyl chloride (PVC) pipes of known inner volume (height 20 cm, diameter 5.25 cm) were used. One end of the pipe was sealed, and the free end of the pipe was inserted to a depth of 2 cm in the soil to make it airtight. Before insertion, a 50-mL beaker containing 40 mL NaOH was placed on a tripod inside the pipes (Pandey et al. 2010). After 24 h, the amount of C[O.sub.2] absorbed by the residual alkali was measured by titration against 0.25 M HCl, using phenolphthalein as an indicator. The C[O.sub.2] output inside the pipes was calculated using following the formula (Witkamp 1966):

m(C[O.sub.2]) = V x N x 22

where m (C[O.sub.2]) is the mass of captured C[O.sub.2](mg), F is the volume of HCl used in titration against saturated NaOH solution (mL) and N is the concentration of HCl. Soil C[O.sub.2] efflux is reported in units of mg C[O.sub.2] [m.sup.-2] [h.sup.-1], whereas and soil C[O.sub.2]-C efflux is reported in units of t [ha.sup.-1] [year.sup.-1].

Measurements of leaf and root litter production

Litter input to the soil at each site was measured through 10 randomly distributed nylon net litter traps (each 50 cm x 50 cm, 15 cm deep), which were placed 50 cm above the ground within a 1-ha area in each site (two traps per permanent plot). Litter from the trap was collected at monthly intervals for 24 months (January 2012-December 2014). The litter material was transported to the laboratory and categorised into tree leaf, wood or branch and miscellaneous, which consisted of mainly tree bark, flowering and fruiting bodies. The amount of the miscellaneous litter category was very small compared with the amount of the other two categories, so it was merged with the wood or branch component and is referred to as the 'non-leaf component' throughout this paper. All litter fractions were ovendried at 80[degrees]C for 48 h and then weighed.

Root biomass of the upper 20 cm soil layer was estimated by excavating 10 randomly selected soil monoliths (10 cm x 10 cm) at least 5 m apart, with two samples from each of the five permanent plots from each site. Seasonal samples were collected (i.e. four times a year) and the roots were recovered after washing over a sieve system. Roots were divided into different diameter classes (<2, 2-5 and 5-10 mm). Root production was calculated as the sum of the difference between the annual maximum and minimum root standing crop of the different diameter classes assuming a single annual pulse of root production (McClaugherty et al. 1982). In ecosystems where distinct troughs and peaks are apparent, such as in cool temperate forests of Japan, this method has been found to be realistic (Tripathi et al. 2005), whereas in the forests where more than one annual peak occurs this method may underestimate net production of fine (<2 mm) roots.

C and N accumulation in the roots was calculated as the product of mean annual mass of different root categories and their total C and N concentrations, whereas C and N input from roots was calculated as the product of the amount of annual root production and their corresponding C and N concentration. Turnover of the root biomass was calculated as the ratio of annual net production to mean annual biomass.

Measurements of leaf and root litter decomposition

In September 2011, mixed tree roots were collected by digging out soil monoliths from the 0-15 cm layer from each site. These roots were washed using a sieve system and oven-dried at 35[degrees]C for 72 h to constant weight. Using callipers, root samples were separated into three diameter categories: fine roots ([less than or equal to]2mm diameter), medium roots (2-5 mm) and coarse roots (5-10 mm). The litter bag technique was used to quantify decomposition rates (Bocock et al. 1960). After adjusting for the initial moisture content, root samples in each category (equivalent to 5 g dry weight) were enclosed in nylon net bags (mesh size 2 mm; 15cmx 15 cm).

In all, 375 bags were prepared (125 bags for each site) for mixed tree species litter and roots of different diameter classes, with 25 bags each for mixed tree leaves, branches and the three root categories. Litter bags containing leaf and wood materials were placed above the soil surface in bunches of five bags in each permanent plot at all sites in the first week of June 2012. Bags containing roots were buried in the soil to a depth of 0-15 cm on the same date in the same manner. Periodically, five bags (one from each bunch) for each category were retrieved at 3-monthly intervals from each study site. Litter bags collected on each sampling date were kept in individual polythene bags and transported to the laboratory. Litter bags were opened and the adhering soil particles were removed and litter samples oven dried at 80[degrees]C for 24 h to constant weight. The dried litter samples were ground and passed through a 2-mm mesh screen for further nutrient analysis.

Chemical analysis

Litterfall samples of different categories collected from the litter traps were pooled for each of the four seasons and powdered. Powdered samples of litterfall for each season and litter categories retrieved periodically (every 3 months) were analysed for C and N concentrations using a Heraeus CHNS/O Elemental analyzer (EuroVector, Mandaluyong City, Philippines) with autosampler (Euro Vector, Model: EuroEA3000) at the Central Instrument Laboratory, Mizoram University.

Initial lignin and cellulose content was determined using a Fibrotron Automatic Fibre Analyser System (Model FRB 6, version 0.1) (Tulin Equipments, Chennai, India). The lignin content of the litter and roots was determined using 1 M [H.sub.2]S[O.sub.4] and cetyltrimethylammonium bromide (CTAB). Weighed samples (0.5 g) were placed into oven-dried sintered glass crucibles. Crucibles, along with the weighed samples, were boiled initially at 350[degrees]C with 100 mL acid detergent solution (ADS) for 10 rnin, followed by boiling at a reduced temperature of 250[degrees]C for 30 min. Then, crucibles were placed in a hot air oven to dry the material, and then weighed. The percentage of lignin content in the material was calculated as follows:

% Lignin = (weight of crucible + acid treated sample) - (weight of empty crucible) x 100/weight of original sample

Cellulose content was determined by treating 0.5-g samples with neutral detergent solution (NDS) and ethoxyethanol and [Na.sub.2]S[O.sub.4]before boiling the samples at 350[degrees]C for approximately 45 min in the Fibrotron Automatic Fibre Analyser System. After boiling, the crucibles were dried in a hot air oven and weighed. After recording the weight, the entire crucible was placed into a muffle furnace at 500[degrees]C for 4-5 h and weighed after cooling. The cellulose content was calculated as follows:

% Cellulose = (weight of crucible + acid treated sample) - (weight of crucible with ash content) / (weight of original sample) x 100

Computations

The mean relative decomposition rate (RDR; mg [g.sup.-1] [day.sup.-1]) was calculated as follows:

RDR = ln([W.sub.1] - [W.sub.0])/([t.sub.1] - [t.sub.0])

where [W.sub.0] is the mass of litter present at time [t.sub.0], [W.sub.1] is the mass of the litter at time [t.sub.1], and [t.sub.1] - [t.sub.0] is the sampling interval (days).

The annual decay constant (k) of litter components was calculated through the negative exponential decay model of Olson (1963) as [x.sub.t]/[x.sub.0]= [exp.sup.(-kt)], where [x.sub.0] is the original mass of litter, [x.sub.t] is the amount of litter remaining after time t, t is the time (year) and k is the decomposition rate ([year.sup.-1]). The time required for 50% and 95% mass loss and nutrient release was calculated as [t.sub.50] = 0.693/k and [t.sub.95] = 3/k respectively.

The nitrogen retranslocation efficiency (NRE; %) of leaves was calculated according to Finzi et al. (2001) as follows:

NRE = [(N in green leaves) - (N in leaf litter)/ (N in green leaves)] x 100

The nitrogen use efficiency (NUE) of leaf was calculated according to Vitousek (1984) as follows:

NUE = LFM/[N.sub.litterfall]

where LFM is litterfall mass (g [m.sup.-1] [year.sup.-1]) and [N.sub.litterfall] is the N content in the litterfall (g [m.sup.-2] [year.sup.-1]).

Statistical analysis

Correlation coefficients were calculated and regression analyses performed between important variables (i.e. monthly rainfall, mean air temperature, humidity, soil moisture, C[O.sub.2] efflux) and litter mass lost during decomposition. Initial litter chemical parameters were also correlated and regressed with litter mass lost. In addition, stepwise multiple regression analysis was performed to determine the effects of abiotic and initial litter chemistry on decomposition rates. One-way analysis of variance (ANOVA) was used to test site-wise differences in the instantaneous decay rate (k). All analyses were conducted using IBM SPSS statistics v20.0 software (IBM, United States). Unless stated otherwise, two-sided P<0.05 was considered significant.

Results

Soil properties and C[O.sub.2] efflux during forest recovery

The soil (0-10 cm) of the study area varied from sandy to sandy loam in texture and was strongly acidic (pH = 4.5-5.1). The mean annual soil properties of the three study sites ranged from 0.57 to 0.87 g [cm.sup.-3] for bulk density, from 21% to 30% for soil moisture content, from 0.98% to 1.96% for total C, from 0.16% to 0.27% for total N, from 1.0 to 2.6 mg [kg.sup.-1] for [P.sub.avail] and from 88 to 210 mg [kg.sup.-1] for [K.sub.exch] (Table 1). Mean annual MBC ranged from 200 to 403 mg [kg.sup.-1], whereas C[O.sub.2] efflux varied from 240 to 389 mg C[O.sub.2] [m.sup.-2] [h.sup.-1] in the three sites. Except for bulk density, values for all physicochemical properties were highest in the NF and lowest in RF-5. Soil C[O.sub.2] efflux was markedly (P< 0.001) seasonal, being lowest in the dry seasons and highest in the wet seasons. Soil C[O.sub.2] efflux was significantly higher in the NF compared with the other sites (Fig. 2).

Seasonal changes in soil C[O.sub.2] efflux rates in the RF-5, RF-15 and NF sites were significantly positively correlated with MBC (r=0.56, 0.80 and 0.88 respectively; (P<0.01), soil moisture (r=0.56, 0.76 and 0.80 respectively; (P<0.01) and soil N (r=0.52, 0.70 and 0.77 respectively; (P<0.01). Stronger correlations were observed between these parameters and C[O.sub.2] efflux in the NF site compared with the RF-15 and RF-5 sites, with weakest correlations seen in the RF-5 site. Soil C[O.sub.2] efflux was significantly negatively correlated (r = -0.60, (P<0.05) with C content in the RF-5 site, whereas it was positively correlated in the RF-15 and NF sites, albeit not significantly. The amount of annual C[O.sub.2]-C efflux in these forests varied significantly (19, 24 and 31 t [ha.sup.-1] [year.sup.-1] in the RF-5, RF-15 and NF sites respectively).

Litter production, N return and NUE during forest recovery

The total annual litterfall for the 2 years (2012 and 2013) in RF-5, RF-15 and NF was in the range 2.66-2.72, 6.47-6.61 and 8.96-9.64 t [ha.sup.-1] [year.sup.-1] respectively (Fig. 3). One-way ANOVA revealed significant (P< 0.05) differences in total annual litterfall among the three forests. In the 2 years, leaf litter comprised 73-77, 75-77 and 71-75% of total litterfall in RF-5, RF-15 and NF respectively, with corresponding values for non-leaf litterfall of 20-22, 17-22 and 20-25%. Of the total non-leaf component, wood litter contributed ~90%; the remainder was miscellaneous litter.

In the 2 years, litterfall was recorded throughout the year, with the bulk (65-75%) occurring during the dry period of the year (November-April) in the three sites. Early litterfall was seen at the regenerating sites compared with the NF site. In both years, litterfall peaked in December in RF-5, compared with peaks in March in RF-15 and NF (Fig. 4).

The leaf litterfall was more concentrated in the short cool--dry period (December-February) of the year, accounting for 53-62% of total leaf litterfall. The remaining litterfall was almost equally distributed in the rainy (18-21%) and summer (19-24%) seasons. Significant negative correlations (r=0.55-0.83, (P <0.01) were obtained between monthly values of leaf litterfall and total rainfall, as well as between mean soil moisture and humidity, and these variables separately accounted for 50-84% variability in the amount of leaf litterfall. Monthly mean air temperature was significantly correlated with litterfall in RF-5, whereas it did not show significant correlations with litterfall in the other two sites.

The total mean biomass of roots (in the upper 20 cm of the soil) varied significantly among the forest sites. Mean root biomass increased significantly with increasing forest age (Table 2). The contribution of fine roots (<2mm) was comparatively higher in the regenerating forests (58-60%) than in the NF (54%). Seasonal variations in the amount of fine root biomass were significant (P< 0.001) among sites.

The total amount of C and N return through litter (leaf, non-leaf and roots) differed significantly (P<0.05) among the RF-5, RF-15 and NF sites (3.23, 4.81 and 5.89 t C [ha.sup.-1] [year.sup.-1] respectively; 0.07, 0.10 and 0.12 t N [ha.sup.-1] [year.sup.-1] respectively; Table 3). The contribution of roots to total C and N return was high in RF-5 and RF-15 compared with NF (36% and 33% vs 29% respective). NUE, NRE and N for leaf litter were comparatively higher in the NF than in the regenerating forest sites. Total litter production in these sites increased significantly (P<0.05) with increasing stand age (7.59, 10.89 and 13.87 t [ha.sup.-1] [year.sup.-1] in RF-5, RF-15 and NF respectively).

Initial chemical composition of litter

The initial chemical composition of different litter material varied considerably among sites (Table 4). In general, leaf litter and small-diameter (<2mm) roots contained higher amounts of cellulose and N. In contrast, branch and higher-diameter roots contained lower amounts of cellulose and N and higher amounts of lignin and C. Significant correlations were observed among chemical quality parameters; for example, cellulose content was negatively correlated with lignin (r=0.78), (P<0.01) and C:N (r = 0.76), (P<0.01) and lignin :N ratios (r = 0.81), (P< 0.001), and positively correlated with N and ash content (r = 0.81), (P < 0.001).

Mass loss during decomposition

Maximum mass loss (30-36%) for different litter components was recorded during the first recovery after 90 days decomposition (Fig. 5). Mass loss at 450 days (as a percentage of initial mass) in the five litter categories was in the order leaf [less than or equal to] roots <2 mm diameter <roots 2-5 mm diameter <roots 5-10 mm diameter <branches. The percentage mass of litter material remaining at the three sites at the end of the study was 17-26% for leaf, 20-25% for roots <2 mm diameter) 34-42% for roots 2-5 mm diameter, 40-46% for roots 5-10 mm diameter and 42-48% for branches (Table 5).

The RDR (4.5-5.6 [mgg.sup.-1] [day.sup.-1]) of leaf and small root litter in the first rainy season was significantly higher (P< 0.05) than the RDR during the winter, summer and second rainy seasons. The same trend was observed for other components with relatively low rates of RDR. The instantaneous annual decay rates (k) in the three sites for different litter categories varied from 0.91-1.26 for leaf and fine roots to 0.55-0.84 for other categories (Table 5). On the basis of k values, the time projections for 50% weight loss varied from 180-236 days for leaf and roots to 291-427 days for other components (Table 5).

Relationship between mass loss and abiotic variables

To assess the effects of prevailing environmental factors on mass loss, we investigated the effects of monthly rainfall (RF), relative humidity (RH), air temperature (AT) and soil moisture (SM). Stepwise multiple regression equations (see below) were developed between mass losses of different litter categories and total RF and the mean of the 3-monthly values of RH, AT and SM between two retrieval dates. Total rainfall and SM explained 50-80% of the variability in the decomposition of different root categories, whereas RF and AT explained 70-80% variability in leaf and branch decomposition.

Leaf = 5.24 + 1.10(RF) - 0.25(AT), P < 0.001, [r.sup.2] = 0.80

Branch = -13.70 + 0.80(RF) - 0.44(AT), P < 0.01, [r.sup.2] = 0.70

Root 0.5 - 2 mm = 11.72 + 0.77(RF) - 0.45(SM), P < 0.001, [r.sup.2] = 0.80

Root 2 - 5mm = 8.57+ 1.16(RF) - 0.58(SM), P < 0.01, [r.sup.2] = 0.70

Root 5 - 10mm = 0.76 + 0.72(RF), P < 0.01, [r.sup.2] = 0.50

Effect of litter quality on mass loss rates

The effect of initial litter quality on decomposition rates was evaluated by correlating mass loss rates of different litter categories at various sampling dates with their initial chemical composition (e.g. cellulose, lignin, ash, C and N content and lignin: N and C : N ratios). The concentrations of cellulose, N and ash content were positively correlated (r = 0.54, 0.64 and 0.65 respectively; (P< 0.05) with mass loss rate in the early (90 days) stages of litter decomposition. Lignin and C: N and lignin :N ratios were significantly negatively correlated (r=-0.59 to -0.81, -0.50 to -0.80 and -0.56 to -0.82 respectively; P <0.05) with mass loss rates at later stages of decomposition.

Changes in C and N stocks in decomposing litter during forest recovery

In the present study, the pattern of C stock loss followed a pattern similar to that of mass loss, reflecting significant decreases during the course of decomposition. However, N stock showed an initial increase followed by a gradual decrease later during successive stages of decomposition (Fig. 6). Increases in N stock during the course of decomposition were significantly greater in RF-5 and RF-15 than in NF for leaf and roots <2 mm, whereas in other components the increase in N stock was not significant across sites. The C concentration in litter categories varied narrowly among different forests, whereas the N concentration in litter varied widely, but not significantly, between the three forests. Among initial litter quality parameters, C, N, lignin and cellulose were found to significantly affect mass loss, C and N mineralisation rates (explaining 90-99% of variability).

Discussion

General pattern of litter input and C and N return during forest recovery

Total annual litterfall in NF (9.7 t [ha.sup.-1] [year.sup.-1]) in the present study lies well within the range reported for tropical forests in India and secondary lowland rain forest of Nigeria (Swamy and Proctor 1994; Odiwe and Muoghalu 2003). However, total litterfall values in the two regenerating forests (2.7 and 6.6 t [ha.sup.-1] [year.sup.-1]) were considerably to slightly lower than the range reported for moist tropical forests in India, secondary lowland rain forest of Nigeria (Odiwe and Muoghalu 2003), tropical rainforest of Australia (Stacker et al. 1995) and northern temperate forest of Japan (Tripathi et al. 2006), which could possibly be due to variations in age and abiotic variables in these sites. The pattern of leaf litterfall was in accordance with total litterfall in these forests, which was broadly comparable with that of the tropical wet and dry evergreen forests of the Indian Western Ghats (Swamy and Proctor 1994; Pragasan and Parthasarathy 2005).

Litterfall in these sites was markedly seasonal and concentrated during the cool and dry period of the year. Austin and Vitousek (2000) reported that in tropical environments the quantity of litter can increase several-fold during the dry season due to increases in leaf litterfall and that its decomposition starts after the onset of the rainy season. This indicates that the litterfall in these forest sites is influenced by the combinations of decreased air temperature, precipitation and soil moisture, which was reflected by the significant negative correlation between litter fall and these variables in the present study. Litterfall in different forests around the world has been reported to be strongly seasonal and driven by air temperature and precipitation (Stacker et al. 1995; Zhou et al. 2006), and the values in the present study are comparable with those of other tropical ecosystems of the world (Austin and Vitousek 2000; Zhou et al. 2006; Zhang et al. 2013).

The early peak leaf litterfall (November-December) in RF-5 compared with RF-15 and NF in the present study reveals that the trees shed leaves earlier to avoid water loss from the canopy under the more rapid soil drying caused by increased temperature, reduced soil moisture and nutrients at RF-5. Recently, Kotowska et al. (2015) reported an increase in temporal fluctuations of litter production as a result of increasing landuse intensity in the tropical forests of Sumatra. In tropical forests, peak litterfall has been reported to be more often associated with several abiotic and biotic factors, such as water stress (Lam and Dudgeon 1985), photoperiod, evapotranspiration (Meentemeyer et al. 1982), low nutrient availability and the percentage of deciduous and semideciduous species (Jordon 1983).

A lower N concentration in leaf litter compared with green leaf indicated higher NUE in the present study (i.e. 83, 87 and 98 in RF-5, RF-15 and NF respectively). NUE in the forests in the present study was towards the lower side of the range reported by Vitousek (1984) for different forests of the world. This is considered to reflect an N-conservation mechanism in plants to develop short-term independence of the N supply from the soil by remobilising the N from permanent parts of trees to new growth during stress. The marginal decrease in NUE in RF-5 is in accordance with the report from Kotowska et al. (2015) in tropical forests of Sumatra with increasing land use intensity.

Litterfall and fine roots are the main source of C and N input to the soil in the forest (Xuluc-Tolosa et al. 2003; Uselman et al. 2007; Ostertag et al. 2008). Studies on inputs from leaves and roots together during secondary succession are limited, because these parameters have been studied separately. Increased aboveground litter input has been reported to reduce the amount of C allocation to fine roots and disproportionately increase soil respiration rates in a moist tropical forest of Central America (Sayer et al. 2007). This will have major implications for C dynamics during land use changes. Marin-Spiotta et al. (2007) have reported that despite marked changes in species composition and litterfall during secondary succession, changes in the process related to C cycling are less pronounced. In the present study, considerable amounts of C (3.2-5.9 t [ha.sup.-1]) and N (69-121 kg [ha.sup.-1]) were added each year through leaf and root litter input in these forest ecosystems with increased contribution of fine roots in the regenerating forests (31-37%) compared with NF (27%). This suggests that fine roots play a more important role than leaf litter in the recovery of C cycling in secondary forests.

Control of litter decomposition kinetics and C and N release during forest recovery

In a variety of forests ecosystems, litter decomposition rates have been reported to be controlled by abiotic and intrinsic litter quality (Zhang et al. 2008; Zhou et al. 2008; Paudel et al. 2015). However, litter decomposition during the course of secondary succession was more strongly affected by site quality than litter quality in secondary forests of Cameroon (Hauser et al. 2005), Puerto Rico (Ostertag et al. 2008) and south-west China (Paudel et al. 2015). Increased temperature and decreased soil moisture content in recovering sites have been reported as possible causes of the decreased rate of decomposition (Martius et al. 2004; Hobbie et al. 2009; Paudel et al. 2015) compared with natural forest, which serves as better microsite for the decomposition of litter (Table 5). In the present study, decomposition kinetics of leaves and fine roots (<2 mm) did not vary significantly in regenerating and natural forest sites, which may be related to changes in abiotic variables and litter quality of two litter types. However, the decomposition rate of roots >2 mm in diameter was significantly lower than that of fine roots and leaves because of the predominant role of initial litter chemistry. Hobbie et al. (2009) reported lack of similarity between fine root and leaf litter decomposition of 11 temperate tree species in south-western Poland, and hypothesised that this may be due, in part, to the role of different traits in affecting the decomposition both above-and belowground. During the process of decomposition, leaves and fine roots (<2mm) showed a distinct tendency towards N immobilisation in developing forests. High immobilisation of N in regenerating forests indicates a tendency to retain N against its probable run-off loss during torrential rains, and this N may be used later by the plants to avoid soil N limitations. Branch litter significantly immobilised N in the NF site, which may help retain N on the forest floor.

The rates of decomposition of different components of litters in the present study were comparable among the three different sites and not significantly different. Similar non-significant results regarding litter decomposition rates were reported among different successional (3, 13 and >50 years) dry forests in Mexico (Xuluc-Tolosa et al. 2003) and successional subtropical wet forests of Puerto Rico (Ostertag et al. 2008), where litter decay rates did not vary with forest age. However, considerably higher litter decomposition rates in the wet compared with dry period suggest that rainfall and its associated variables (like soil moisture) play a vital role in the process of decomposition in these forests by enhancing the activity of decomposer organisms compared with other climatic variables (Austin and Vitousek 2000; Xuluc-Tolosa et al. 2003). The decreased rates of C[O.sub.2] effluxes during the winter months, as noted in the present study, may be due to decreased microbial populations and activity, which was also reflected in the decreased MBC value.

The period of rapid mass loss rates during litter decomposition in roots could have been caused by chemical characteristics of the roots or by some attribute of the belowground decomposition environment (Ostertag and Hobbie 1999). Thus, tissue chemistry appears to be a good predictor of both leaf and root decomposition rates, although their ability to retain N varied among the three sites, suggesting some site-specific requirements. The below ground environment has been cited as more conducive to rapid mass loss because of moisture conditions that promote either microbial activity or more rapid colonisation of decomposing material by microbes (Ostertag and Hobbie 1999). In the present study, mass loss rates of leaves and fine roots did not vary significantly between the sites, but the tendency for N immobilisation differed among the sites, reflecting functional changes in the ecosystems during the course of decomposition. Significantly lower decomposition rates in coarse than fine roots may be related to the morphology and the initial quality of these roots.

The present study is based on a field experiment pertaining to litter input and decomposition in two secondary forest sites and a natural forest, and true replicates of the sites could not be established due to constraints such as difficulties with measurement and limited resources. Although all possible care has been taken to validate our data so that it is comparable with other field experiments and to explain statistically significant differences between the sites, readers should keep in mind the absence of site replication while interpreting our results.

Conclusion

The findings of the present study demonstrate forest litter production and decomposition both above- and belowground. In particular, belowground litter in regenerating forests enhanced C and nutrient cycling processes, which can recover rapidly (in less than two decades) during the course of forest recovery in this region, as reflected by enhanced fine root production and turnover in younger forests compared with the NF site. This suggests that ecosystem functions in secondary forests can be restored despite variations in species composition and site condition compared with natural forests. The species differences in litter quality may have little effect on litter decomposition rates, but the rainfall and its associated variables may have a strong effect. Recovering forests developed ecosystem-level strategies like early leaf fall to reduce soil water loss through transpiration, the addition of a proportionately greater amount of organic matter to the soil from fine roots and conservation of N through immobilisation during decomposition. All these strategies indicate a significant role for litter in accelerating C and N cycling properties in young secondary forests during forest recovery so that they may potentially provide ecosystem services similar to those of a natural forest in a short period of time (i.e. less than two decades).

https://doi.org/10.1071/SR16265

Conflicts of interest

The authors declare no conflicts of interest.

Acknowledgements

The authors thank Julian Brasington and Hillary Ford. Bangor University, for their comments on an earlier draft of the manuscript. The authors also thank the anonymous reviewers for their critical comments that improved the quality of the manuscript. The University Grants Commission is gratefully acknowledged for financial support. The authors also thank the Department of Forestry, Mizoram University, for providing laboratory facilities.

References

Aide TM, Grau HR (2004) Globalization, migration, and Latin American ecosystems. Science 305, 1915-1916. doi:10.1126/science.ll03179

Anderson JM, Ingram ISI (1993) 'Tropical soil biology and fertility: a handbook of methods.' 2nd edn. (CAB International: Wallingford, UK)

Austin AT, Vitousek PM (2000) Precipitation, decomposition and litter decomposability of Metrosideros polymorpha in native forests on Hawai. Journal of Ecology 88. 129-138. doi: 10.1046/j. 1365-2745.2000.00437.x

Bocock KL, Gilbert D. Capstick CK, Twin DC, Ward JS, Woodman MJ (1960) Changes in leaf litter when placed on the surface of the soils with contrasting humus types I. Losses in dry weight of oak and ash leaf litter. Journal of Soil Science 11, 1-9. doi: 10.1111/j.1365-2389.1960.tb02196.x

Boyero L, Cardinale BJ, Bastian M, Pearson RG (2014) Biotic vs. abiotic control of decomposition: a comparison of the effects of simulated extinctions and changes in temperature. PLoS One 9. e87426. doi: 10.1371/journal.pone.0087426

Celentano D. Zahawi RA, Finegan B. Ostertag R, Cole RJ, Holl KD (2011) Litterfall dynamics under different tropical forest restoration strategies in Costa Rica. Biotropica 43, 279-287. doi:10.1111/j.1744-7429.2010.00688.x

Champion HG, Seth SK (1968) 'A revised survey of the forest types of India.'(Government of India: New Delhi, India)

Chazdon RL (2014) 'Second growth: the promise of tropical forest regeneration in an age of deforestation.'(University of Chicago Press: Chicago, IL, USA)

Chazdon RL, Letcher SG, van Breugel M, Martinez-Ramos M. Bongers F, Finegan B (2008) Rates of change in tree communities of secondary neotropical forests following major disturbances. Philosophical Transactions of the Royal Society of London B Biological Sciences 362, 273-289.

Chen H, Gurmesa GA, Liu L, Zhang T. Fu S (2014) Effects of litter manipulation on litter decomposition in a successional gradients of tropical forests in southern China. PLoS One 9, e99018. doi:10.1371/journal.pone.0099018

Denslow JS, Guzman GS (2000) Variation in stand structure, light and seedling abundance across a tropical moist forest chronosequence, Panama. Journal of Vegetation Science 11, 201-212. doi: 10.2307/3236800

Economics and Statistics Department Government of Mizoram (2014) 'State Statistical handbook of Mizoram.' (Economics and Statistics Department Government of Mizoram: North East India).

Feldpausch TR, Rondon MA, Femande ECM, Riha SJ, Wandelli E (2004) Carbon and nutrient accumulation in secondary forests regenerating on pastures in Central Amazonia. Ecological Applications 14, 164-176. doi: 10.1890/01-6015

Finzi AC, Allen AS, DeLucia EH, Ellsworth DS, Schlesinger WH (2001) Forest litter production, chemistry and decomposition following two years of free-air C[O.sub.2] enrichment. Ecology 82, 470-484.

Fioretto A, Papa S, Fuggi A (2003) Litterfall and litter decomposition in a low Mediterranean shrubland. Biology and Fertility of Soils 39. 37-14. doi: 10.1007/s00374-003-0675-5

Gavazov K, Mills R, Spiegelberger T, Lenglet J, Buttler A (2014) Biotic and abiotic constraints on the decomposition of Fagussvlvatica leaf litter along an altitudinal gradient in contrasting land-use types. Ecosystems 17, 1326-1337. doi: 10.1007/s10021-014-9798-9

Hauser S, Gang E, Norgrove L, Birang MA (2005) Decomposition of plant material as an indicator of ecosystem disturbance in tropical land use systems. Geoderma 129, 99-108. doi: 10.1016/j.geoderma.2004.12.037

Hobbie SE, Oleksyn J, Eissenstat DM, Reich PB (2009) Fine root decomposition rates do not mirror those of leaf litter among temperate tree species. Oecologia 162. 505-513.

Holl KD (2002) Tropical moist forest restoration. In 'Handbook of ecological restoration', Vol. II. (Eds A. J. Davy and M. Perrow.) pp. 539-558. (Cambridge University Press: Cambridge, UK)

Jordon CF (1983) Productivity of tropical rain forest ecosystems and the implications for their useas future wood and energy sources. In 'Tropical rainforests ecosystems-structure and function'. (Ed. F. B. Golley) pp. 117-136. (Elsevier Scientific Publishing: Oxford, UK)

Kotowska MM, Leuschner C, Triadiati T, Hertel D (2015) Conversion of tropical lowland forest reduces nutrient return through litterfall, and alters nutrient use efficiency and seasonality of net primary production. Oecologia 180. 601-618. doi: 10.1007/s00442-015-3481-5

Lam PKS, Dudgeon D (1985) Seasonal effects on litterfall in a Hong Kong forest. Journal of Tropical Ecology 1, 55-64. doi: 10.1017/S0266467400000079

Lambers H, Raven JA, Shaver GR, Smith SE (2008) Plant nutrient-acquisition strategies change with soil age. Trends in Ecology & Evolution 23, 95-103. doi: 10.1016/j.tree.2007.10.008

Lewis SL, Edwards DP, Galbraith D (2015) Increasing human dominance of tropical forests. Science 349. 827-832. doi: 10.1126/science.aaa9932

Loaiza-Usuga JC, Leon-Pelaez JD, Gonzalez-Hernandez MI, Gallardo-Lancho JF, Osorio-Vega W, Correa-Londono G (2013) Alterations in litter decomposition patterns in tropical montane forests of Colombia: a comparison of oak forests and coniferous plantations. Canadian Journal of Forest Research 43, 1-6.

Marin-Spiotta E, Ostertag R, Silver WL (2007) Long-term patterns in tropical reforestation: plant community composition and aboveground biomass accumulation. Ecological Applications 17, 828-839. doi:10.1890/06-1268

Martin PH, Sherman RE, Fahey TJ (2004) Forty years of tropical forest recovery from agriculture: structure and floristics of secondary and old-growth riparian forests in the Dominican Republic. Biotropica 36. 297-317.

Martius C, Hofer H, Garcia MVB, Rombke J, Hanagarth W (2004) Litter fall, litter stocks and decomposition rates in rainforest and agroforestry sites in central Amazonia. Nutrient Cycle in Agroecosystems 68, 137-154.

Mtambanengwe F, Kirchman H (1995) Litter from tropical savanna woodland (Miombo): chemical composition and C and N mineralization. Soil Biology & Biochemistry 11, 1639-1651. doi: 10.1016/0038-0717(95)00073-N

McClaugherty CA, Aber JD, Melillo JM (1982) The roles of fine roots in organic matter and nitrogen budgets of forest ecosystems. Ecology 63, 1481-1490. doi: 10.2307/1938874

Meentemeyer V, Box EO, Thompson R (1982) World patterns and amounts of terrestrial plant litter production. Bioscience 32, 125-128. doi: 10.2307/1308565

Odiwe Al, Muoghalu JI (2003) Litterfall dynamics and forest floor litter as influenced by fire in a secondary lowland rain forest in Nigeria. Tropical Ecology 44, 241-248.

Olson JS (1963) Energy storage and the balance of producers and decomposers in ecological systems. Ecology* 44, 322-331. doi: 10.2307/1932179

Ostertag R, Hobbie SE (1999) Early stages of root and leaf decomposition in Hawaiian forests: effects of nutrient availability. Oecologia 121. 564-573. doi: 10.1007/s004420050963

Ostertag R, Marin-Spiotta E, Silver WL, Schulten J (2008) Litterfall and decomposition in relation to soil carbon pools along a secondary forest chronosequence in Puerto Rico. Ecosystems 11. 701-714. doi:10.1007/s10021-008-9152-1

Page AL, Miller RL, Keeny DR (Eds) (1982) "Methods of soil analysis. Part 2, second edition.' (American Society of Agnonomy, Soil Science Society of America: Madison, WI, USA)

Pandey RR, Sharma G, Tripathi SK, Singh AK (2007) Litterfall, litter decomposition and nutrient dynamics in a subtropical natural oak forest and managed plantation in northeastern India. Forest Ecology and Management 240. 96-104. doi:10.1016/j.foreco.2006.12.013

Pandey RR, Sharma G, Singh TB. Tripathi SK (2010) Factors influencing soil C[O.sub.2] efflux in a north eastern Indian oak forest and plantation. African Journal of Plant Science 4. 280-289.

Pandey S, Sheikh GA, Bhat AH (2016) Dynamics of litterfall in nutrient cycling and forest preservation. International Journal of Multidisciplinaiy Research 2, 2455-3662.

Paudel E, Dossa GGO, de Blecourt M, Beckschafer P. Xu J, Harrison RD (2015) Quantifying the factors affecting leaf litter decomposition across a tropical forest disturbance gradient. Ecosphere 6, 1-20. doi: 10.1890/ES15-00112.1

Pragasan AL, Parthasarathy N (2005) Litter production in tropical dry evergreen forests of south India in relation to season, plant life-forms and physiognomic groups. Current Science 88. 1255-1263.

Purahong W, Schloter M, Pecyna MJ, Kapturska D, Daumlich V, Mital S, Buscot F, Hofrichter M, Gutknecht JLM, Kruger D (2015) Uncoupling of microbial community structure and function in decomposing litter across beech forest ecosystems in central Europe. Scientific Reports 4. doi: 10.1038/srep07014

Roy PS, Roy A (2010) Land use and land cover change in India: a remote sensing and GIS perspective. Journal of the Indian Institute of Science 90, 489-502.

Sayer EJ, Powers JS, Tanner TVJ (2007) Increased litterfall in tropical forest boosts the transfer of soil C[O.sub.2] to the atmosphere. PLoS One 2, e1299. doi: 10.1371/journal.pone.0001299

Silver WL, Kueppers LM, Lugo AE. Ostertag R, Matzek V (2004) Carbon sequestration and plant community dynamics following reforestation of tropical pasture. Ecological Applications 14, 1115-1127. doi: 10.1890/03-5123

Singh ShB, Mishra BP, Tripathi SK (2015) Recovery of plant diversity and soil nutrients during stand development in subtropical forests of Mizoram, northeast India. Biodiversitas 16, 205-212. doi: 10.13057/biodiv/d160216

Slik JWF, Paoli G, McGuire K, Amaral I, Barroso J, Bastian M, Blanc L (2013) Large trees drive forest aboveground biomass variation in moist lowland forests across the tropics. Global Ecology and Biogeography 22, 1261-1271. doi: 10.1111/geb.12092

Stocker GC, Thompson WA, Irvine AK, Fitzsim JD, Thomas PR (1995) Annual patterns of litterfall in a lowland and table- land rainforest in tropical Australia. Biotropica 27, 412-420.

Swamy HR, Proctor J (1994) Litterfall and nutrient cycling in four rainforests in the Sringeri area of the Indian Western Ghats. Global Ecology and Biogeography Letters 4, 155-165. doi: 10.2307/2997533

Tripathi SK, Singh KP (1992a) Nutrient immobilization and release pattern during plant decomposition in a dry tropical bamboo savanna, India. Biology and Fertility of Soils 14, 191-199. doi:10.1007/BF00346060

Tripathi SK, Singh KP (19926) Abiotic and litter quality control during the decomposition of different plant parts in dry tropical bamboo savanna in India. Pedobiologia 36, 109-124.

Tripathi SK, Sumida A, Shibata H, Uemura S, Ono K, Hara T (2005) Growth and substrate quality of fine roots and soil nitrogen availability in a young Betula ermanii forest of Northern Japan: effects of the removal of understory dwarf bamboo (Sasa kurilensis). Forest Ecology and Management 212, 278-290. doi:10.1016/j.foreco.2005.03.030

Tripathi SK, Sumida A, Shibata H, Ono K, Uemura S, Kodama Y, Hara T (2006) Leaf litterfall and decomposition of different above- and belowground parts of birch (Betula ermanii) trees and dwarf bamboo (Sasa kurilensis) shrubs in a young secondary forest in Northern Japan. Biology and Fertility of Soils 43, 237-246. doi: 10.1007/s00374-006-0100-y

Uselman SM. Quails RG, Lilienfein J (2007) Fine root production across a primary successional ecosystem chronosequence at Mt. Shasta, California. Ecosystems 10, 703-717. doi: 10.1007/s10021-007-9045-8

van Breugel M, Bongers F. Marti'Nez-Ramos M (2007) Species dynamics during early secondary forest succession: recruitment, mortality and species turnover. Biotropica 39, 610-619. doi: 10.1111/j.1744-7429.2007.00316.x

Vance ED. Brookes PC, Jenkinson DS (1987) An extraction method for measuring soil microbial biomass carbon. Soil Biology & Biochemistry 19, 703-707. doi: 10.1016/0038-0717(87)90052-6

Vitousek PM (1984) Litterfall, nutrient cycling, and nutrient limitation in tropical forests. Ecology 65. 285-298. doi:10.2307/1939481

Walker J, Reddel P (2007) Retrogressive succession and restoration on old landscapes. In 'Linking restoration and ecological succession'. (Eds L. R. Walker, J. Walker, and R. J. Hobbs.) pp. 1-16. (Springer Science: New York, NY, USA)

Witkamp M (1966) Rates of carbon dioxide evolution from the forest floor. Ecology 47, 492-494. doi: 10.2307/1932992

Xuluc-Tolosa FJ, Vester HFM, Ramirez-Marcial N, Castellanos-Albores J, Lawrence D (2003) Leaf litter decomposition of tree species in three successional phases of tropical dry secondary forest in Campeche, Mexico. Forest Ecology and Management 174, 401--412. doi: 10.1016/S0378-1127(02)00059-2

Zhang D, Hui D, Luo Y, Zhou G (2008) Rates of litter decomposition in terrestrial ecosystems: global patterns and controlling factors. Journal of Plant Ecology 1, 85-93. doi:10.1093/jpe/rtn002

Zhang K, Cheng X, Dang H, Ye C, Zhang Y, Zhang Q (2013) Linking litter production, quality and decomposition to vegetation succession following agricultural abandonment. Soil Biology & Biochemistry 57, 803-813. doi: 10.1016/j.soilbio.2012.08.005

Zhou GY, Guan LL, Wei XH, Zhang DQ, Zhang QM (2006) Litterfall production along successional and altitudinal gradients of subtropical monsoon evergreen broadleaved forests in Guangdong, China. Plant Ecology 188, 77-89. doi: 10.1007/s 11258-006-9149-9

Zhou G, Guan L, Wei X, Tang X, Liu S, Liu J, Zhang D, Yan J (2008) Factors influencing leaf litter decomposition: an intersite decomposition experiment across China. Plant and Soil 311. 61-72. doi:10.1007/s11104-008-9658-5

C. Lalnunzira (A) and S. K. Tripathi (A,B)

(A) Department of Forestry, School of Earth Sciences and Natural Resources Management, Mizoram University, Aizawl, Tanhril -796004, India.

(B) Corresponding author. Email: sk_tripathi@rediffmail.com

Received 3 October 2016, accepted 10 November 2017, published online 16 March 2018

Caption: Fig. 1. Climatic data for the study area. Data were collected from the Weatherstation, Department of Environmental Sciences, Mizoram University, Tanhril, India.

Caption: Fig. 2. Seasonal changes in C[O.sub.2] efflux in three forests during the study period. Data are the mean [+ or -] s.e.m. (n = 3). RF-5, 5-year-old regenerating forest; RF-15, 15-year-old regenerating forest; NF, and natural forest.

Caption: Fig. 3. Total annual litterfall during the study period in two regenerating and one natural forest. Data are the mean [+ or -] s.e.m. Different letters indicate significant differences among the forests. RF-5, 5-year-old regenerating forest; RF-15, 15-year-old regenerating forest; NF, and natural forest.

Caption: Fig. 4. Monthly litterfall in the three forests during the period 2012-14. Data are the mean [+ or -] s.e.m. (n=10). RF-5, 5-year-old regenerating forest; RF-15, 15-year-old regenerating forest; NF, and natural forest.

Caption: Fig. 5. Mass remaining plotted against the number of days after litter placement for (a) leaves, (b) branches, (c) fine roots ([less than or equal to]2mm diameter), (d) medium roots (2-5 mm diameter) and (e) coarse roots (5-10 mm diameter) in the study sites.RF-5, 5-year-old regenerating forest; RF-15, 15-year-old regenerating forest; NF, and natural forest.

Caption: Fig. 6. Temporal changes in the amount of (a-e) carbon (C) and (f-j) nitrogen (N) stock in different litter components during the course of decomposition. (a,f) Leaves, (b,g) branches, (c, h) fine roots ([less than or equal to]2 mm diameter), (d, i) medium roots (2-5 mm diameter) and (e,j) coarse roots (5-10 mm diameter). RF-5, 5-year-old regenerating forest; RF-15, 15-year-old regenerating forest; NF, and natural forest.
Table 1. Soil physicochemical properties (in the 0-10 cm layer)
in the study plots

Data are given as the mean [+ or -] s.e.m. of six periodic
measurements at each site (n = 18). Within rows, values with
different letters differ significantly (P< 0.05). RF-5, 5-year-old
regenerating forest; RF-15, 15-year-old regenerating forest; NF,
and natural forest; MBC, microbial biomass carbon

Soil parameters                           RF-5

pH (1:2.5 w/v [H.sub.2]O)           5.1 [+ or -] 1.1a
Bulk density(g [cm.sup.-3])        0.87 [+ or -] 0.03a
Moisture content (%)               21.6 [+ or -] 2.6a
Total N (%)                        0.16 [+ or -] 0.04a
Organic C (%)                      0.98 [+ or -] 0.04a
Soil C : N ratio                   4.04 [+ or -] 0.01a
Available P (mg [kg.sup.-1])       1.25 [+ or -] 2.02a
Exchangeable K (mg [kg.sup.-1])      88 [+ or -] 12a
MBC (mg [kg.sup.-1])                210 [+ or -] 15a
C[O.sub.2] efflux (mgC[O.sub.2]     246 [+ or -] 24a
[m.sup.-2] [h.sup.-1])

Soil parameters                          RF-15

pH (1:2.5 w/v [H.sub.2]O)          4.8 [+ or -] 0.8b
Bulk density(g [cm.sup.-3])       0.69 [+ or -] 0.04ab
Moisture content (%)              24.9 [+ or -] 2.6a
Total N (%)                       0.20 [+ or -] 0.06a
Organic C (%)                     1.37 [+ or -] 0.04b
Soil C : N ratio                  4.89 [+ or -] 0.12b
Available P (mg [kg.sup.-1])       1.4 [+ or -] 2.4a
Exchangeable K (mg [kg.sup.-1])    163 [+ or -] 18b
MBC (mg [kg.sup.-1])               275 [+ or -] 22ab
C[O.sub.2] efflux (mgC[O.sub.2]    299 [+ or -] 20b
[m.sup.-2] [h.sup.-1])

Soil parameters                           NF

pH (1:2.5 w/v [H.sub.2]O)         4.52 [+ or -] 0.67c
Bulk density(g [cm.sup.-3])       0.57 [+ or -] 0.05b
Moisture content (%)              30.7 [+ or -] 3.1b
Total N (%)                       0.27 [+ or -] 0.03b
Organic C (%)                     1.96 [+ or -] 0.05c
Soil C : N ratio                   5.2 [+ or -] 0.1c
Available P (mg [kg.sup.-1])       2.6 [+ or -] 3.1b
Exchangeable K (mg [kg.sup.-1])    210 [+ or -] 13c
MBC (mg [kg.sup.-1])               403 [+ or -] 21c
C[O.sub.2] efflux (mgC[O.sub.2]    390 [+ or -] 22c
[m.sup.-2] [h.sup.-1])

Table 2. Root biomass (g [m.sup.-2]) in the different root
categories up to 20 cm soil depth in the two regenerating
sites and the natural forest

Data are given as the mean [+ or -] s.e.m. of four seasons
(winter, summer, monsoon and post-monsoon; n = 20). Within
columns, values with different letters differ significantly
(P<0.05). RF-5, 5-year-old regenerating forest; RF-15,
15-year-old regenerating forest; NF, and natural forest

                               Root
                          categories (mm)
Site          <2                2-5

RF-5    190 [+ or -] 26    86 [+ or -] 11
RF-15   300 [+ or -] 35   131 [+ or -] 15
NF      374 [+ or -] 46   177 [+ or -] 22

Site         5-10              Total

RF-5     50 [+ or -] 9    324a [+ or -] 35
RF-15    69 [+ or -] 8    500b [+ or -] 50
NF      138 [+ or -] 13   689c [+ or -] 52

Table 3. Litter dry matter, C and N return to soil through
litterfall and root mortality, and nitrogen use efficiency
(NUE) in two regenerating and one natural forest site

RF-5, 5-year-old regenerating forest; RF-15, 15-year-old
regenerating forest; NF, and natural forest; NRE,
N retranslocation efficiency

                                Leaf litter and
                                   fine roots

                               RF-5   RF15   NF

Leaf mass return (A)           412    578    668
(g [m.sup.-2] [year.sup.-1])

Root mass return (B)           231    285    281
(g [m.sup.-2] [year.sup.-1])

Total litter mass return       643    863    949
(g [m.sup.-2] [year.sup.-1])

Total C return                 270    374    433
(g [m.sup.-2] [year.sup.-1])

Total N return                 6.3    8.5    9.5
(g [m.sup.-2] [year.sup.-1])

Leaf NUE                        83     89    96

Leaf NRE                        30     38    44

                                Non-leaf litter
                                and coarse roots

                               RF-5   RF15    NF

Leaf mass return (A)            65    170    280
(g [m.sup.-2] [year.sup.-1])

Root mass return (B)            51     56     66
(g [m.sup.-2] [year.sup.-1])

Total litter mass return       106    226    306
(g [m.sup.-2] [year.sup.-1])

Total C return                  53    107    156
(g [m.sup.-2] [year.sup.-1])

Total N return                 0.72   1.81   2.57
(g [m.sup.-2] [year.sup.-1])

Leaf NUE

Leaf NRE

                                      Total
                               RF-5   RF15     NF

Leaf mass return (A)           477     748    1030
(g [m.sup.-2] [year.sup.-1])

Root mass return (B)           282     341     357
(g [m.sup.-2] [year.sup.-1])

Total litter mass return       759    1089    1284
(g [m.sup.-2] [year.sup.-1])

Total C return                 323     481     589
(g [m.sup.-2] [year.sup.-1])

Total N return                 6.9    10.3    12.08
(g [m.sup.-2] [year.sup.-1])

Leaf NUE

Leaf NRE

(A) Including grasses. Grass contributed ~40% to total mass
return in RF-5, which decreased sharply in RF-15 and NF.

(B) Root mass return in the 0-20 cm layer was calculated as
the product of diameter class (<2,2-5 and 5-10mm) root
production and root turnover values.

Table 4. Initial chemical composition of
v branches and roots in in two regenerating
and one natural forest

Where appropriate, data are given as the mean [+ or -]
s.e.m. (n = 3). RF-5, 5-year-old regenerating forest;
RF-15, 15-year-old regenerating forest;
NF, and natural forest

Site    Litter       Cellulose         Lignin
        component    (mg [g.sup.-1])   (mg [g.sup.-1])

RF-5    Leaf         386 [+ or -] 12   286 [+ or -] 13
        Branch       230 [+ or -] 7    397 [+ or -] 9
        Roots (mm)
          <2         326 [+ or -] 11   273 [+ or -] 15
          2-5        298 [+ or -] 6    360 [+ or -] 10
          5-10       245 [+ or -] 10   392 [+ or -] 11
RF-15   Leaf         403 [+ or -] 14   267 [+ or -] 6
        Branch       242 [+ or -] 11   362 [+ or -] 8
        Roots (mm)
          <2         332 [+ or -] 14   284 [+ or -] 11
          2-5        325 [+ or -] 9    344 [+ or -] 10
          5-10       293 [+ or -] 15   371 [+ or -] 15
NF      Leaf         413 [+ or -] 18   254 [+ or -] 8
        Branch       264 [+ or -] 10   375 [+ or -] 14
        Roots (mm)
          <2         366 [+ or -] 12   284 [+ or -] 2
          2-5        344 [+ or -] 19   320 [+ or -] 1
          5-10       310 [+ or -] 9    350 [+ or -] 2

Site    Litter       Carbon                 Nitrogen
        component    (mg [g.sup.-1])     (mg [g.sup.-1])

RF-5    Leaf         437 [+ or -] 4      12 [+ or -] 2
        Branch       444 [+ or -] 7     7.1 [+ or -] 0.4
        Roots (mm)
          <2         382 [+ or -] 3       7 [+ or -] 2
          2-5        395 [+ or -] 5     7.1 [+ or -] 0.6
          5-10       439 [+ or -] 7     6.5 [+ or -] 0.4
RF-15   Leaf         442 [+ or -] 5      11 [+ or -] 4
        Branch       452 [+ or -] 8       7 [+ or -] 2
        Roots (mm)
          <2         381 [+ or -] 10    8.4 [+ or -] 1.2
          2-5        426 [+ or -] 9       7 [+ or -] 1
          5-10       469 [+ or -] 12    6.3 [+ or -] 0.5
NF      Leaf         448 [+ or -] 6      10 [+ or -] 5
        Branch       466 [+ or -] 12      7 [+ or -] 3
        Roots (mm)
          <2         395 [+ or -] 8    10.5 [+ or -] 0.7
          2-5        430 [+ or -] 5     8.2 [+ or -] 0.4
          5-10       458 [+ or -] 3     7.0 [+ or -] 0.5

Site    Litter       C: N    Lignin: N     Ash content
        component    ratio     ratio     (mg [g.sup.-1])

RF-5    Leaf         36.62     23.96          47.10
        Branch       63.43     56.43          24.50
        Roots (mm)
          <2         51.34     36.69          39.23
          2-5        55.40     50.49          36.10
          5-10       66.52     59.39          32.73
RF-15   Leaf         40.98     24.51          51.07
        Branch       61.75     49.45          24.40
        Roots (mm)
          <2         45.20     33.69          39.60
          2-5        57.26     46.24          34.83
          5-10       64.42     50.96          29.03
NF      Leaf         43.54     24.77          54.33
        Branch       63.49     51.09          29.10
        Roots (mm)
          <2         37.78     27.89          40.30
          2-5        52.50     44.51          38.30
          5-10       65.24     49.86          33.70

Table 5. Decomposition parameters for mass loss and time required
for 50% and 95% decay ([t.sub.50] and [t.sub.95] respectively)

RF-5, 5-year-old regenerating forest; RF-15,15-year-old
regenerating forest; NF, and natural forest

Site    Litter       Mass remaining    Annual
        component     at 450 days      decay
                     (% of initial)   rate (k)

NF      Leaf             17.82          1.26
        Branch           42.60          0.66
        Roots (mm)
          <2             20.03          1.16
          2-5            34.32          0.84
          5-10           40.33          0.69
RF-15     Leaf           22.47          1.01
        Branch           47.07          0.61
        Roots (mm)
          <2             26.75          0.91
          2-5            40.20          0.69
          5-10           45.85          0.64
RF-5    Leaf             26.03          0.98
        Branch           48.20          0.55
        Roots (mm)
          <2             25.83          1.01
          2-5            42.67          0.61
          5-10           46.76          0.58

Site    Litter       [t.sub.50]   [t.sub.95]
        component      (days)       (days)

NF      Leaf           180.80       782.68
        Branch         365.50      1582.24
        Roots (mm)
          <2           193.91       839.45
          2-5          291.60      1262.34
          5-10         343.45      1486.80
RF-15     Leaf         208.86       904.14
        Branch         413.81      1791.39
        Roots (mm)
          <2           236.49      1023.79
          2-5          342.20      1481.40
          5-10         399.91      1731.22
RF-5    Leaf           231.68      1002.94
        Branch         427.30      1849.79
        Roots (mm)
          <2           230.40       997.41
          2-5          366.13      1584.97
          5-10         410.24      1775.93
COPYRIGHT 2018 CSIRO Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2018 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Lalnunzira, C.; Tripathi, S.K.
Publication:Soil Research
Article Type:Report
Geographic Code:9INDI
Date:May 1, 2018
Words:10958
Previous Article:Does 3,4-dimethylpyrazole phosphate or N-(n-butyl) thiophosphoric triamide reduce nitrous oxide emissions from a rain-fed cropping system?
Next Article:Change in soil organic carbon and nitrogen stocks eight years after conversion of sub-humid grassland to Pinus and Eucalyptus forestry.
Topics:

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |