J For Res (2008) 13:25–34 DOI 10.1007/s10310-007-0044-6
ORIGINAL ARTICLE
Leaf litter decomposition of dominant tree species of Namdapha National Park, Arunachal Pradesh, northeast India Atiqur Rahman Barbhuiya Æ Ayyanadar Arunachalam Æ Prabhat Chandra Nath Æ Mohammed Latif Khan Æ Kusum Arunachalam
Received: 20 October 2006 / Accepted: 5 June 2007 / Published online: 12 December 2007 The Japanese Forest Society and Springer 2007
Abstract Rates of weight loss and nutrient (N and P) release patterns were studied in the leaf litter of the dominant tree species (Ailanthus grandis, Altingia excelsa, Castanopsis indica, Duabanga sonneriatioides, Dysoxylum binectariferum, Mesua ferrea, Shorea assamica, Taluma hodgsonii, Terminalia myriocarpa and Vatica lancefolia) of a tropical wet evergreen forest of northeast India. Nitrogen and phosphorus mineralization rate and decay pattern varied significantly from species to species. In general, the decay pattern, characterized by using a composite polynomial regression equation, exhibited three distinct phases of decay during litter decomposition—an initial slow decay phase (0.063% weight loss day-1), followed by a rapid decay phase (0.494% weight loss day-1) and a final slow decay phase (0.136% weight loss day-1). The initial chemical composition of the litter affected decomposition rates and patterns. Species like D. sonneriatoides, D. binectariferum, and T. hodgsonii with higher N and P content, lower carbon and lignin content, and lower C:N ratio and lignin:N ratio exhibited relatively faster decomposition rates than the other species, for example M. ferrea, C. indica and A. grandis. A slow decay rate was recorded for species such as M. ferrea, C. indica, and A. grandis. The initial N and P content of litter showed significant positive correlations with decay rates. Carbon and
A. R. Barbhuiya A. Arunachalam P. Chandra Nath M. Latif Khan K. Arunachalam Department of Forestry, North Eastern Regional Institute of Science and Technology, Nirjuli 791109, Arunachal Pradesh, India A. R. Barbhuiya (&) Department of Forestry, Mizoram University, Aizawl 796009, Mizoram, India e-mail:
[email protected]
lignin content, lignin:N, and C:N showed significant negative correlations with decay rates. Soil total N and P, and rainfall, soil temperature, and soil moisture had positive correlations with decay rates. The rapid decomposition rates observed in comparison with other different forest litter decay rates confirm that tropical wet evergreen forest species are characterized by faster decomposition rates, indicating a faster rate of organic matter turnover and rapid nutrient cycling. Keywords Decay rate Lignin Litter decomposition Nitrogen Phosphorus Tropical wet evergreen forest
Introduction The term ‘‘decomposition’’ is defined as the process of biological disintegration of dead organic materials whereby mineralization of complex organic compounds into simple inorganic forms takes place. This loss of nutrients from decomposing litter is a key process governing the availability of nutrients in ecosystems (Moore et al. 2006). The release of nutrients from forest litter through natural decomposition processes is recognised as an important part of the plant biomass that is made available for further plant growth (Maclean and Wein 1978). The importance of litter decomposition to nutrient cycling and ecosystem function has long been known, and a huge number of studies has been undertaken (Gillon et al. 1999; Moore et al. 2006), many in response to regional or species-specific concerns. Attempts to predict the effects of climate change have recently prompted studies of litter from wider ranges of plant types (Perez-Harguindeguy et al. 2000). The effect of litter quality is obvious from the different decay rates of various tissue types, but identifying the
123
26
Study site The study was conducted in a tropical wet evergreen forest ([200 years old approximately) in the Namdapha National Park (27230 3000 N to 27390 4000 N latitude to 96150 200 E to 96580 3300 E longitude) in the Changlang district of Arunachal Pradesh, northeast India. The total park area is 1,985 km2 over an elevation range of 250–4,571 m asl, of which 177 km2 is buffer zone. The region experiences four seasons—winter (mid-November to February), spring (March to April), monsoon (May to September), and a brief autumn (October to mid-November). The major perennial rivers flowing through Namdapha National Park include the Noa-Dihing, Deban, Namdapha, and Burma Nala. In addition to these, innumerable seasonal rainfed streams and streamlets also drain the park. The annual rainfall ranged from 2,000 to 4,300 mm between 1994 and 2004. An experimental plot of 100 m 9 100 m was established inside the Namdapha National Park (27260 15.800 – 27260 19.700 N latitude to 96280 12.300 –96280 10.400 E longitude). The vegetation in the plot was dominated by mixed
123
1000
45
900
40
800
35
700
30
600 25 500 20 400 15
300 200
10
100
5
0
0 Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar
Soil moisture content (%) & temperature(°C)
tropical evergreen species. The experimental plot was on a plain located at an altitude of 350 m asl with a mean annual temperature of 21C. Average monthly soil temperature and moisture content were 31C and 38.7%, respectively, during summer and 16C and 22.9%, respectively, during winter (Fig. 1). Vegetation, microclimate, and soil
Rainfall (mm)
particular litter characteristics that are consistently and closely related to decomposability has proven surprisingly difficult. Across a broad range of litter type C:N ratio seems to be the best predictor of decay rate (Perez-Harguindeguy et al. 2000), while the lignin content or lignin:N ratio is better correlated with decay rates (Loranger et al. 2002). Long-term studies have indicated that the factors that best correlate with rates of early decay are often not the same as those that relate to long-term decay (Joffre et al. 2001; Yang and Janssen 2002). Berg and Staaf (1980) and Sangha et al. (2006) suggested that early decomposition is regulated by nutrient concentrations (especially N and P) whereas the late-stage decay is regulated by lignin concentration. Different species have different nutrient release patterns, which are related to quality, season, and environmental factors (Abiven et al. 2005; Arunachalam et al. 2003). Namdapha National Park has a variety of tree species, but there is too little information on the relative rates of decay of foliar litter of these species and litter chemical parameters useful for predicting their decay rates. Thus, in this study we have quantified the decomposition rates and nutrient mineralization patterns of the dominant tropical tree species Ailanthus grandis, Altingia excelsa, Castonopsis indica, Dysoxylum binectariferum, Duabanga sonneriatioides, Mesua ferrea, Taluma hodgsonii, Terminalia myriocarpa, Shorea assamica, and Vatica lancefolia of Namdapha National Park, Arunachal Pradesh, India, to understand their role in the organic matter and nutrient turnover that has a bearing on the overall productivity of the ecosystem.
J For Res (2008) 13:25–34
Months
Fig. 1 Monthly variation of rainfall, temperature (filled triangles), and moisture content (open squares) of the study area
Table 1 Vegetation, microclimate and soil (0–15 cm) physicochemical characteristics of the Namdapha National Park Parameters
Values
Vegetation Density (plants ha-1) Trees
610
Shrubs
13,280
Herbs
17,400
Basal area (m2 ha-1) Trees Shrubs
98.6 3.8
Herbs
0.424
Leaf litter fall (kg ha-1 year-1)
4896 ± 101
Soil properties Texture Sand (%)
58.4 ± 3.1
Silt (%)
19.8 ± 1.1
Clay (%)
21.8 ± 0.1
Texture class
Sandy clay loam
Water-holding capacity (g g-1) -3
Bulk density (g cm )
0.665 ± 0.003 0.691 ± 0.032
Moisture content (g g-1)
0.314 ± 0.001
pH (1:2.5 w/v H2O)
5.2 ± 0.08
Organic C (g kg-1)
18.00 ± 1.00
Total N (g kg-1)
3.00 ± 0.20
C:N P concentration (g kg-1)
6.0 ± 0.001 0.30 ± 0.01
± standard error from five replicate measurements
J For Res (2008) 13:25–34
27
characteristics have been summarized in Table 1. Tree density and basal area were 610 trees ha-1 and 98.6 m2 ha-1, respectively, in the selected plot. The ten dominant tree species in the experimental plot are listed in in Table 2. Soil was sandy clay loam, acidic (pH = 5.2), and the C:N ratio was 6 (Table 1).
Methodology Leaf litter chemistry Freshly fallen leaves of the ten dominant tree species, viz., Ai. grandis, A. excelsa, C. indica, D. binectariferum, Table 2 List of tree species for decomposition study of Namdapha National Park Name of the species
Family
Growth characteristics
Ailanthus grandis Prain.
Simarubaceae
Medium
Altingia excelsa Noron.
Hamamelidaceae
Slow
Castanopsis indica (Roxb.) Miq.
Fabaceae
Slow
Duabanga sonneratioides Buch.
Sonneratiaceae
Fast
Dysoxylum binectariferum Hk.f ex Bed
Meliaceae
Medium
Mesua ferrea L. Shorea assamica Dyer.
Clusaceae Dipterocarpaceae
Slow Slow
Talauma hodgsonii Hk. F. & Thom.
Magnoliaceae
Slow
Terminalia myriocarpa Muell.
Combretaceae
Fast
Vatica lancefolia (Roxb.) Blume.
Dipterocarpaceae
Slow
D. sonneriatoides, M. ferrea, T. hodgsonii, T. myriocarpa, S. assamica, and V. lancefolia, were collected during the peak litter fall period (March). The litter samples were air dried in the laboratory and sub-samples were kept at 80C for 48 h to determine the dry mass. The oven-dried materials were powdered in a Wiley mill to pass through 1 mm pore size stainless-steel mesh and analysed for their chemical composition. The ash content was determined by igniting 1 g ground litter sample at 550C for 6 h in a muffle furnace and a total of 50% of the ash-free mass was calculated as the carbon (C) content (Allen et al. 1974). Soil organic carbon (SOC) was determined by the complete oxidation method (Nelson and Sommers 1975). Nitrogen (N) was estimated by use of a semi-micro Kjeldahl procedure entailing acid digestion, distillation, and titration, in accordance with Anderson and Ingram (1993). Total phosphorus (P) was determined by triacid digestion, followed by colorimetric reaction with ammonium and stannous chloride (molybdenum blue method; Jackson 1958). Lignin, cellulose, and hemicellulose content were determined gravimetrically according to Allen et al. (1974).
Litter decomposition Decomposition of leaf litter of dominant tree species was studied in the forest stand using a nylon bag (15 cm 9 15 cm) technique (Gilbert and Bocock 1960). The mesh size was 2 mm, small enough to prevent major losses of litter samples yet large enough to permit aerobic microbial activity and free entry of small soil organisms. Air-dried material (5 g) was placed in each bag, which was then stitched with nylon thread. For each species 80 bags were prepared and randomly dispersed on the experimental forest floor in the month of March 2003. After 30, 60, 90,
Table 3 Initial chemical composition of leaf litter of the dominant tree species of the Namdapha National Park Species
C (mg g-1)
N (mg g-1) P (mg g-1)
Lignin (mg g-1) C:N
L:N
Cellulose (mg g-1)
Hemicellulose (mg g-1)
A. grandis
363.3 ± 26.8 10.6 ± 0.2
0.502 ± 0.011 170.3 ± 23.7
34.3 ± 1.3 16.1 ± 1.0 234.8 ± 36.8 61.6 ± 0.8
A. excelsa
341.5 ± 2.5
14.9 ± 0.6
0.513 ± 0.031 248.0 ± 8.8
23.1 ± 2.6 16.8 ± 2.0 320.2 ± 30.1 55.7 ± 4.6
C. indica
285.0 ± 9.7
6.9 ± 0.3
0.713 ± 0.017 180.0 ± 10.0
41.9 ± 3.2 26.5 ± 2.5
D. binectariferum 302.7 ± 19.3 18.7 ± 3.1
0.645 ± 0.001 157.9 ± 11.6
21.0 ± 1.1
8.4 ± 0.7 395.8 ± 36.1 49.5 ± 8.5 6.7 ± 1.0 384.0 ± 29.7 60.2 ± 5.7
28.6 ± 2.1
70.3 ± 5.0
D. sonneriatoides
303.3 ± 51.0 20.5 ± 2.0
0.914 ± 0.024 137.6 ± 13.8
18.8 ± 1.6
M. ferrea
281.7 ± 12.2
0.456 ± 0.006 248.5 ± 26.4
49.4 ± 4.5 43.6 ± 2.6 219.0 ± 14.7 78.6 ± 6.3
5.7 ± 0.8
T. hodgsonii
315.1 ± 21.3 13.2 ± 1.0
0.856 ± 0.001 210.6 ± 31.4
23.9 ± 2.9 16.0 ± 1.2 265.6 ± 45.3 40.8 ± 3.0
T. myriocarpa
301.6 ± 20.0 15.3 ± 3.3
0.474 ± 0.021 187.5 ± 17.6
19.7 ± 2.1 12.3 ± 1.0 241.4 ± 37.0 63.8 ± 7.0
S. assamica
315.0 ± 11.7 10.7 ± 4.5
0.617 ± 0.003 234.0 ± 18.8
29.4 ± 2.5 21.9 ± 1.3 275.1 ± 26.3 60.0 ± 1.9
V. lancefolia
340.1 ± 26.6 13.6 ± 2.3
0.543 ± 0.020 223.4 ± 21.4
25.0 ± 1.1 16.4 ± 2.9 225.6 ± 33.0 95.6 ± 9.3
± standard error from five replicate measurements L:N, lignin:nitrogen ratio
123
28
J For Res (2008) 13:25–34
120, 150, 180, 210, 240, 270, 300, 330, 360, and 390 days five-litter bags for each species were brought to the laboratory, carefully avoiding loss of materials from the bags. The litter bags were washed in a bucket full of tap water, by swirling briefly, and carefully decanted through a 2-mm mesh size sieve to remove extraneous matter. The litter was then dried at 80C for 48 h and weighed. The samples were powdered and used for analysis of N and P concentrations. The annual decay constant (k) was calculated by following the negative exponential decay model (Olson
Fig. 2 Decomposition of leaf litter of different tree species. Vertical lines represent standard error from five replicate measurements
120
1963): k = -ln (X/X0)/t, where, X0 is the initial dry weight, X is the dry weight remaining at the end of the investigation, and t is the time period. The time required for 50% (t50) and 99% (t99) weight loss was calculated as t50 = 0.693/k and t99 = 5/k (Olson 1963). The effects on the rate of decomposition of the litter of climatic variables, initial litter chemistry, and a few soil characteristics were assessed by using a simple linear regression function, Y = a + bx (Zar 1974). ANOVA was used to test for differences among different species,
M. ferrea
A. grandis
100 Y=98.56+0.33X-2.43X2-0.15X3 r=0.997, P
