Scholarly article on topic 'High carbon stocks in roadside plantations under participatory management in Bangladesh'

High carbon stocks in roadside plantations under participatory management in Bangladesh Academic research paper on "Biological sciences"

CC BY-NC-ND
0
0
Share paper
Academic journal
Global Ecology and Conservation
OECD Field of science
Keywords
{"Basal area based model" / "Carbon sequestration" / "Climate change" / Livelihood / REDD+ / "Trees outside forest"}

Abstract of research paper on Biological sciences, author of scientific article — Md. Mizanur Rahman, Md. Enamul Kabir, A.S.M. Jahir Uddin Akon, Kazuo Ando

Abstract Plantations are important REDD+strategies for increasing carbon sequestration while enhancing local livelihoods. Reforestation along roads and highways under participatory forest management in southwestern Bangladesh could contribute to REDD+. This study assessed the diversity and structure of roadside plantations in order to develop a basal area based generalized allometric model for estimating above- and below-ground tree biomass carbon in Southwestern Bangladesh. All woody plants with d.b.h. ⩾ 2 cm were identified and their diameters measured in 108 systematically selected zigzag plots of equal size ( 2 × 10 m ). A total of 36 species in 17 families were recorded. Leguminosae accounted for 28% of species and 94% of the total estimated biomass carbon. We estimated a mean stem density of 4528 ha − 1 , basal area of 52.6 m 2 ha − 1 and biomass carbon of 192.80 Mg ha−1. Samanea saman, Dalbergia sissoo, Acacia nilotica, and Leucaena leucocephala accounted for most density, basal area, and carbon. We developed and validated three allometric models with equal strength (R2 0.94–0.98) using generalized linear regression. Roadside plantations in Bangladesh can now surely participate in the UNFCCC’s carbon mitigation and adaptation mechanism, but challenges to their long-term sustainability must be addressed.

Academic research paper on topic "High carbon stocks in roadside plantations under participatory management in Bangladesh"

Contents lists available at ScienceDirect

Global Ecology and Conservation

journal homepage: www.elsevier.com/locate/gecco

Original research article

High carbon stocks in roadside plantations under participatory management in Bangladesh

CrossMaik

Md. Mizanur Rahmana, Md. Enamul Kabirb-d'*, A.S.M.Jahir Uddin Akonc, Kazuo Andod

a Remote Sensing Division, Center for Environmental and Geographic Information Services, House 6, Road23/C, Gulshan 1, Dhaka 1212, Bangladesh

b Forestry and Wood Technology Discipline, Khulna University, Khulna 9208, Bangladesh c Bangladesh Forest Department, Banabhaban, Agargaon, Dhaka-1207, Bangladesh

d Center for Southeast Asian Studies, Kyoto University, 46 Shimoadachi-cho, Yoshida Sakyo-ku, Kyoto 606-8501, Japan

article info

abstract

Article history:

Received 6 January 2015

Received in revised form 29 January 2015

Accepted 29 January 2015

Available online 4 February 2015

Keywords:

Basal area based model Carbon sequestration Climate change Livelihood REDD+

Trees outside forest

Plantations are important REDD+strategies for increasing carbon sequestration while enhancing local livelihoods. Reforestation along roads and highways under participatory forest management in southwestern Bangladesh could contribute to REDD+. This study assessed the diversity and structure of roadside plantations in order to develop a basal area based generalized allometric model for estimating above- and below-ground tree biomass carbon in Southwestern Bangladesh. All woody plants with d.b.h. >2 cm were identified and their diameters measured in 108 systematically selected zigzag plots of equal size (2 x 10 m). A total of 36 species in 17 families were recorded. Leguminosae accounted for 28% of species and 94% of the total estimated biomass carbon. We estimated a mean stem density of 4528 ha-1, basal area of 52.6 m2 ha-1 and biomass carbon of 192.80 Mgha-1. Samanea saman, Dalbergia sissoo, Acacia nilotica, and Leucaena leucocephala accounted for most density, basal area, and carbon. We developed and validated three allometric models with equal strength (R2 0.94-0.98) using generalized linear regression. Roadside plantations in Bangladesh can now surely participate in the UNFCCC's carbon mitigation and adaptation mechanism, but challenges to their long-term sustainability must be addressed. © 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction

Global warming and biodiversity loss are the two important currently debated issues among the world's scientists and policy makers (Zhang et al., 2011), caused mainly by fossil fuel burning and deforestation during the last few decades (van der Werf et al., 2009). The last century finished with an increase in global temperature by 0.74 °C and the atmospheric CO2 concentration of 379 ppm (UNFCCC, 2007; IPCC, 2013). Furthermore, atmospheric carbon dioxide would be doubled by 2050 if the current rate of increase continues and will lead to the global temperature rise of up to 2-4 °C (IPCC, 2013). A projection by IPCC (2013) revealed that by the end of 21st century the global sea level will rise by 28-98 cm due to melting of polar ice, which would badly alter low-lying coastal countries (e.g. Bangladesh, Maldives, The Netherlands) existence and livelihoods pattern. Forests retention, coupled with various reforestation and afforestation programmes, tropical in particular, can play

* Corresponding author at: Forestry and Wood Technology Discipline, Khulna University, Khulna 9208, Bangladesh. Tel.: +880 1776300400; fax: +880 41724717.

E-mail address: menamulkabir@gmail.com (M.E. Kabir). http://dx.doi.org/10.1016/j.gecco.2015.01.011

2351-9894/© 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/ licenses/by-nc-nd/4.0/).

an important role in mitigating global climate change through sequestering atmospheric carbon (see Dixon et al., 1994a,b; Jose, 2009; Kumar, 2011).

Forests in Africa, Latin America, South and Southeast Asia, and the Pacific region have been experiencing the highest carbon emissions due to deforestation (FAO, 2010). In the least developed countries like Bangladesh, 62% of total carbon emissions originate from deforestation (IPCC, 2007). The Kyoto Protocol, the main instrument of the UNFCCC, has introduced the CDM concept among the low-income people who can store carbon through changes in land use patterns (Roshetko et al., 2007; Takimoto et al., 2008). Under the current arrangements, reduced emissions from deforestation and forest degradation, and enhancing forest carbon stocks in developing countries (REDD+) does not consider smallholder trees, but only large scale forests. However, small patches of trees outside forests, if not cut, can contribute towards reducing emissions of carbon to the atmosphere. The only reason why smallholder trees are not a major focus under the REDD+ arrangement is because their contribution towards carbon sequestration is not well documented (Nair, 2012). To improve country negotiations for REDD+ and other carbon market mechanisms, it is necessary to provide evidence about the potential contribution of smallholder trees outside forests to biodiversity conservation, livelihood options and carbon sequestration (Nair, 2012; Jashimuddin and Inoue, 2012).

Yet, globally attention has been given to what extent managed landscapes, such as agroforests, community forests, village woodlots and roadside plantations under participatory management, could hold carbon and contribute to climate change mitigation (see Albrecht and Kandji, 2003; Roshetko et al., 2007; Saha et al., 2010; Kumar and Nair, 2011; Nair, 2012). The global coverage of agroforests is 1023 million ha which represents a carbon sequestration potential of 1.9 Pg of carbon over 50 years at a rate of 94 Mg ha-1 in managed landscapes (Dixon et al., 1994a,b; Nair et al., 2009).

Participatory forest management has been practiced in Bangladesh since 1976. Several pilot projects from 1982 to 1987 provided the experience for launching the countrywide participatory forest management project in the forms of restored forests, agroforests, village woodlots, and road and highway plantations in 1989 (Kabir and Webb, 2005). Increasing the supply of forest products, especially fuel wood, to improve rural socioeconomic conditions and reversing the process of environmental degradation through proper soil and water conservation are the main objectives of participatory forest management in Bangladesh. As an outcome, 48,420 ha of roadside plantations, 30,666 ha of woodlots and 8778 ha of agroforestry plantations have been raised during the last 30 years in Bangladesh (Jashimuddin and Inoue, 2012). Members of the poorer sections, with special preference to landless, land-poor and rural women of the surrounding rural communities, are targeted as participants in the roadside plantations. The participants who protect the plantations have the right to harvest and consume or market all intermediate products in the forms of leaves, twigs and dead branches for household fuel. According to Forest Department policy, the participants have a right to receive a pre-determined share of 40% of the receipts from auction of the trees in a local market after the rotation (usually 10-12 years). Yet, commonly practiced roadside plantation under participatory management in Bangladesh has so far received no research attention in estimating its potential contribution to livelihood supplementation and carbon sequestration.

Species-level tree biomass carbon estimation using diameter at breast height (dbh) with a tree density based allometric model is becoming popular (Pandey et al., 2014; Rahman et al., 2014). However, for quick calculation of biomass carbon, a basal area based allometric model could be another important option as both basal area and biomass carbon have a strong relation to dbh (Torres and Lovett, 2012; Rahman et al., 2014). Studies have made significant contributions in estimating ecosystem level aboveground carbon stocks using basal biomass (see Torres and Lovett, 2012; Rahman et al., 2014). The present focus of REDD+ is examining to what extent carbon sequestration through forest restoration and plantation establishment is related to biodiversity conservation, poverty reduction, and carbon sequestration. Therefore, this study aimed to develop generalized allometric equations based on basal area to estimate tree biomass carbon content of the roadside plantations under participatory management in southwestern Bangladesh.

2. Methods

2.1. Study area

Southwestern Bangladesh is primarily a floodplain landmass lying between 21.50° and 23.91°N latitude and 88.55° and 90.35°E longitude (Fig. 1). The region is bounded by India in the west and the Bay of Bengal in the south. Excessive siltation and sedimentation in the upper stream has resulted in phenomenal floods in the plain land areas of the region during the monsoon and severe drought during the dry season. Regular floods and droughts cause enormous damage to the lives and property in the region. The deltaic landscape of this region is a primarily low (<10 m above a.s.l.), flat, and fertile plain. The coastal plain is partly sandy and saline, and varies from 1 to 15 km in width (Kabir and Webb, 2008). Calcareous to non-calcareous soils and peat are the basic soil types extending over the study area. Coastal regions have some areas with acid sulphate and peat soils (Kabir and Webb, 2008).

Thirty-six percent of the study area is cultivable, 9% uncultivable, 37% littoral mangrove forestland, and 18% un-surveyed (BBS, 2013). Rice, wheat, jute, sugarcane, pulses, and potatoes are the principal agricultural crops from the cultivable lands (Kabir and Webb, 2008). Various types of vegetables, spices, fruits, and nuts are also important cultivated crops. There are no primary forests in the study area except the inaccessible littoral Sundarbans mangrove forest. Intensive shrimp (tiger prawn) culture is a newly emerging economic activity along the coastal regions of southwest Bangladesh (Kabir and Webb, 2008).

Fig. 1. Study area. Bagerhat, Satkhira, Jessore, Narail and Jhenaidah districts in southwestern Bangladesh.

Fig. 2. Photographs of roadside plantation in southwestern Bangladesh. Tender plantation along both the mud (upper left) and paved (lower left) roads. Beneficiary of the plantation (upper right) and the harvested products (lower right) after the rotation for benefit sharing among stakeholders.

All the districts in southwestern Bangladesh adjoin the capital of the country with a relatively good communication network. Highway and railway lines connect all the districts of the region. Paved roads connect each sub-district with the adjoining district headquarters. The majority of the roads within sub-district boundaries are mud with slowly developing paved roads. The majority of rural roads are muddy and become unusable during the rainy season. The Local Government and Engineering Department (LGED) have been converting rural mud roads into paved roads, but the progress is slow. In rural areas, communication facilities are developing rapidly under the rural public works programme. Almost all roads and highways in southwestern Bangladesh have been planted on both sides with the active participation of the local community and local government under the social forestry programme of the Bangladesh Forest Department (see Fig. 2).

A tropical to subtropical monsoon climate characterizes the region, marked with seasonal variations, moderately warm temperatures, heavy rainfall, and excessive humidity (Kabir and Webb, 2008). Three distinct seasons—summer (March-May), rainy (June-October), and winter (November-February) - are characteristic of the region (Kabir and Webb, 2008). The mean annual temperature is 26 °C (range: 19-32 °C). In some places temperatures go down to 7 °C during the winter and reach up to 40 °C or more during the summer (Kabir and Webb, 2008). Total rainfall during the monsoon accounts for 80% of the total annual rainfall. The annual average rainfall of the study area is 1800 ± 268 mm, ranging from 1400 to 2600 mm (Kabir and Webb, 2008). The potential evapotranspiration is 65-129 mm. The annual average relative humidity of the region is 78% (Kabir and Webb, 2008). March is the least humid month (65%) while humidity during the monsoon (June-September) is 75% (Kabir and Webb, 2008).

2.2. Sampling design

A total of 108 plots of equal size (2 x 10 m) were selected following systematic sampling in a zigzag manner (Fig. 3) on both the sides of road, in order to reduce microsite variation. Successive plots were 200 m apart for model development (Fig. 3(a) and 500 m apart for model validation (Fig. 3(b)). Plots were laid out in two slots in same year plantations (2005-2006). At first in Bagerhat district, 77 rectangular plots were laid out along the roadside plantation to collect data to develop a basal area based allometric model. Secondly, 31 plots were laid out along the roadside plantation in four other districts in the region, 10 from Satkhira, 8 fromJessore, 7 from Narail and 6 fromJhenaidah districts (Fig. 1), in order to validate the basal area based allometric model. The data from all 108 plots were later used to assess the floristic composition, stand structure, species diversity and biomass carbon.

a 10m 200 m 200 m 10 m

b Plot- 1 Plot- 3 id

<N IN

*■ Road

Plot - 2 £ (N

b 10m 500 m 500 m 10m

fc Plot - I Plot - 3 fc

fN <N

Plot - 2 s IN

Fig. 3. Schematic diagram of the zigzag plot layout (a) for model development and (b) for model validation along the roadside plantations in southwestern Bangladesh.

2.3. Data collection

All woody plants with dbh > 2 cm, were censused from all sample plots of the roadside plantations. In each sample plot, every individual tree was identified and recorded to species level. The diameters of all identified trees were measured at breast height (1.3 m above ground) using a diameter tapeTMand recorded accordingly.

2.4. Data analysis

2.4.1. Species diversity and structure

Species diversity H' (Shannon and Weaver, 1949), evenness J' (Pielou, 1977) and richness R(Margalef, 1958) indices were computed using BASIC program SPDIVERS.BAS (Ludwig and Reynolds, 1988). The importance value index was computed using the derivatives of density, basal area, and frequency of each recorded species in characterizing the stand structure (see Zhang et al., 2005). ANOVA was used to test the differences between species diversity, richness and evenness, biomass carbon, basal area and tree density of the roadside plantations across study sites. The Least Significant Difference (LSD) test was performed for multiple comparisons if any significant difference was found between species diversity, richness and evenness, biomass carbon, basal area and tree density of the roadside plantations across the study sites.

2.4.2. Tree biomass and carbon

Tree biomass estimation from species- and site-specific allometric models is tedious using the destructive harvest method, particularly in tropical and subtropical regions, because of the presence of numerous species and individuals in multiple layers. Common allometric equations that have been developed by felling trees in different tropical and subtropical regions of the world are an easier alternative. Taking this point into account, the allometric equation for aboveground carbon estimation of Chave et al. (2005) was employed, as it covers a wide geographical and diameter range of vegetation of all types. Below ground biomass and carbon were estimated using the regression model suggested by Cairns et al. (1997) as the most cost effective and practical method of determining root biomass.

The wood density data were obtained from the World Agroforestry Database (Carsan et al., 2012) and the Global Wood Density Database (Chave et al., 2009; Zanne et al., 2009). The relatively poor physical condition of the roadside plantations and their isolation from natural systems usually provides approximately 20% less total biomass than what natural systems provide (Aguaron and McPherson, 2012). Estimated biomass using Chave et al. (2005) and Cairns et al. (1997) allometric equation was therefore multiplied by 0.80 to calculate total biomass as suggested by Aguaron and McPherson (2012). Finally, the total biomass was multiplied by 0.5 to compute actual tree carbon content as 50% of wood's total biomass is considered to be carbon. A Generalized Linear Regression Model (GLRM) was employed to develop and validate the basal area based carbon estimation model.

3. Results

3.1. Species diversity and structure

A total of 36 species in 17 families were recorded from 108 sample plots (Table 1). Of the 36 species, Samanea saman, Dalbergia sissoo, Acacia nilotica, and Leucaena leucocephala were the most important species considering stem density, basal area, and biomass carbon content (Table 1). Leguminosae was the most dominant family, accounting for 28% of all the recorded species and 94% of the total estimated biomass carbon (Table 2). A total of 978 individuals were recorded in the

Table 1

Relative abundance, relative frequency, relative coverage, importance value, and biomass carbon content of trees in roadside plantations in southwestern Bangladesh. List according to the biomass carbon content (from high to low).

Species RD RF RC IVI% BCC%

Samanea saman (Jacq.) Merr. 16.56 30.12 17.77 21.48 27.01

Acacia nilotica Karst. 32.52 17.58 13.55 21.22 24.37

Dalbergia sissoo Roxb. 18.61 22.71 19.58 20.30 23.82

Leucaena leucocephala (Lam.) de Wit 12.07 14.71 9.64 12.14 12.68

Melia azedarach L. 1.84 2.31 4.22 2.79 2.28

Gmelina arborea Roxb. 1.84 1.41 3.61 2.29 0.78

Swietenia macrophylla King 1.64 0.48 3.92 2.01 0.20

Acacia auriculiformis A. Cunn. ex Benth. 1.33 1.60 3.01 1.98 1.28

Albizia procera (Roxb.) Benth. 0.72 2.05 1.81 1.52 2.56

Spondias pinnata (L.f.) Kurz 1.33 0.30 2.71 1.45 0.11

Trewia polycarpa Benth. & Hook.f. 1.33 1.50 1.20 1.34 0.97

Azadirachta indica A.Juss. 1.74 0.32 1.51 1.19 0.25

Artocarpus heterophyllus Lam. 1.02 0.20 1.81 1.01 0.08

Terminalia arjuna Wight & Arn. 0.72 0.36 1.81 0.96 0.30

Senna siamea (Lam.) Irw. & Barneby 0.92 1.15 0.60 0.89 0.77

Albizia richardiana King & Prain 0.61 0.45 1.51 0.86 0.35

Phyllanthus emblica L. 0.51 0.26 1.20 0.66 0.25

Bombax ceiba L. 0.41 0.49 0.90 0.60 0.31

Mangifera indica L. 0.92 0.24 0.60 0.59 0.12

Tectona grandis L.f. 0.41 0.10 1.20 0.57 0.06

Pithecellobium dulce (Roxb.) Benth. 0.41 0.44 0.60 0.48 0.60

Psidium guajava L. 0.31 0.03 0.90 0.41 0.02

Polyalthia longifolia (Sonn.) Hook.f. & Thomson 0.31 0.03 0.90 0.41 0.02

Aegle marmelos (L.) Correa 0.20 0.23 0.60 0.35 0.27

Ziziphus nummularia (Burm.f.) W. & A. 0.20 0.18 0.60 0.33 0.19

Morus alba L. 0.20 0.14 0.60 0.32 0.06

Tamarindus indica L. 0.20 0.03 0.60 0.28 0.02

Lagerstroemia speciosa (L.) Pers. 0.20 0.03 0.60 0.28 0.01

Anthocephalus chinensis (Lmk.) A. Rich. ex Walp. 0.20 0.18 0.30 0.23 0.08

Ficus religiosa L. 0.10 0.21 0.30 0.20 0.12

Moringa oleifera Lam. 0.10 0.08 0.30 0.16 0.01

Trema orientalis (L.) Blume 0.10 0.05 0.30 0.15 0.02

Alstonia scholaris (L.) R. Br. 0.10 0.04 0.30 0.15 0.02

Ficus benghalensis L. 0.10 0.01 0.30 0.14 0.00

Annona reticulata L. 0.10 0.00 0.30 0.14 0.00

Syzygium cumini (L.) Skeels 0.10 0.00 0.30 0.14 0.00

Table 2

Relative abundance, relative frequency, relative coverage, importance value, and biomass carbon content of plant families in roadside plantations in southwestern Bangladesh. List according to the biomass carbon content (from high to low).

SL. no Family RD RF RC IVI BCC%

1 Leguminosae 83.95 52.94 90.84 75.91 93.47

2 Meliaceae 5.21 12.75 3.10 7.02 2.73

3 Verbenaceae 2.25 7.35 1.51 3.70 0.84

4 Anacardiaceae 2.25 4.90 0.54 2.56 0.23

5 Euphorbiaceae 1.84 3.92 1.75 2.51 1.21

6 Moraceae 1.43 4.90 0.56 2.30 0.27

7 Combretaceae 0.72 2.94 0.36 1.34 0.30

8 Myrtaceae 0.41 1.96 0.04 0.80 0.02

9 Annonaceae 0.41 1.96 0.03 0.80 0.02

10 Bombacaceae 0.41 1.47 0.49 0.79 0.31

11 Rutaceae 0.20 0.98 0.23 0.47 0.27

12 Rhamnaceae 0.20 0.98 0.18 0.46 0.19

13 Lythraceae 0.20 0.98 0.03 0.40 0.01

14 Rubiaceae 0.20 0.49 0.18 0.29 0.08

15 Moringaceae 0.10 0.49 0.08 0.23 0.01

16 Ulmaceae 0.10 0.49 0.05 0.21 0.02

17 Apocynaceae 0.10 0.49 0.04 0.21 0.02

sample plots. No significant difference was observed in species richness, diversity, and evenness across the five study sites (P < 0.05; Table 3).

An average density of 4528 stems ha-1 (range: 3000-7950), dbh of 12.45 cm (range: 8.02-12.67) and basal area of 52.6 m2 ha-1 (range: 20.7-95.3) were recorded from roadside plantations across five study sites (Table 4). Multiple comparisons with LSD found no significant differences in stem density and diameter across the study sites (P < 0.05), while

Table 3

Mean species diversity, evenness and richness indices of roadside plantations across five study sites in southwestern Bangladesh. Figures in parenthesis are standard errors.

Species parameters Study sites Average

Bagerhat Jessore Jhenaidah Narail Satkhira

Diversity (H') 1.26(0.08) 1.28(0.42) 1.25(0.35) 1.01 (0.30) 0.98(0.22) 1.22(0.07)

Evenness (J') 0.75(0.04) 0.81(0.26) 0.79(0.22) 0.64(0.19) 0.62(0.14) 0.74(0.04)

Richness (R) 1.00(0.07) 1.22(0.41) 1.13(0.34) 0.88(0.26) 0.68(0.19) 0.98(0.07)

Table 4

Stem density, diameter and basal area of roadside plantations across five study sites in southwestern Bangladesh. Figures in parenthesis are standard errors.

Parameter Study sites Average

Bagerhat Jessore Jhenaidah Narail Satkhira

Density (ha-1) 4487(362) 3000(532) 3000(675) 3143(705) 7950(626) 4528(297)

Diameter (cm) 11.59(0.43) 12.67(1.21) 8.02(1.30) 12.51 (1.35) 11.67(0.36) 11.45(0.35)

Basal area (m2 ha-1) 51.9(3.2) 44.5(13.8) 20.7(8.6) 43.2(10.1) 95.3(6.1) 52.6(3.0)

Table 5

Pearson correlation analysis showing relationship of aboveground, belowground, and biomass carbon with species diversity, evenness, and richness of roadside plantations across five study sites in southwestern Bangladesh. Figures in parenthesis are P values.

Carbon level Pearson correlation (r)

Species diversity (H') Species evenness (J') Species richness (R)

Aboveground 0.03(0.77) 0.03(0.75) -0.06(0.52)

Belowground 0.03 (0.72) 0.03 (0.74) -0.07 (0.50)

Total biomass 0.03(0.76) 0.03(0.75) -0.06(0.52)

Bagerhat Jessore Jhenaidah Narail Satkhira Mean Study sites

Fig. 4. Carbon and biomass content of the roadside plantations across five study sites in southwestern Bangladesh.

basal area was significantly different (P > 0.05). Satkhira district accounted for the highest density and basal area while Jhenaidah district accounted for the least (Table 4).

3.2. Tree biomass and carbon content

An average biomass carbon of 192.80 Mg ha-1 (range: 56.75-380.11) was calculated from the roadside plantations (Fig. 4); 86% above ground and 14% below ground. Above- and below-ground biomass carbon content was significantly different across the five study sites (P < 0.05). The LSD results for multiple comparisons revealed that the Satkhira site has the highest biomass carbon content and Jhenaidah has the lowest. However, no significant difference was found in biomass carbon between Bagerhat, Jessore and Narail districts (P > 0.05). Sites with higher basal area and stem density tend to store more carbon. Pearson's correlation analyses of plot level aboveground, belowground, and biomass carbon content with species diversity, evenness and richness indices showed no significant relationships (P > 0.05; Table 5).

Fig. 5. Basal area based biomass models at the left side of this study validated with models established by Chave et al. (2005) and Cairns et al. (1997) at the right side.

3.3. Basal area based allometric models and their validation

Three types of models were developed for carbon assessment from the plot level mean basal area (Eqs. (1)-(3)). We found a strong (mean R2 = 0.96; for Linear 0.94, Polynomial 0.94 and Power 0.98 models) and significant (P < 0.05) relationship between mean biomass carbon and mean basal area for roadside plantations (Fig. 5). All three models were tested against 31 plots with Chave et al. (2005) and Cairns et al. (1997) for validation. Results of the GLRM revealed that all three models showed significant (P < 0.05) and strong relationships (R2 = 0.84) with established models (Chave et al., 2005; Cairns et al., 1997) based on biomass carbon content (Fig. 5). Given the high regression R2 (range: 0.94-0.98), our three models are equally strong in calculating the tree biomass carbon content. Therefore our basal area based allometric models are equally suitable for calculating biomass carbon content from the trees in open stands.

Biomass C = 4.061 x BA - 22.516

Biomass C = -9.188 + 3.501 x BA + 0.005 x BA2

Biomass C = 1.239 x BA1360.

(1) (2) (3)

4. Discussion

Controlling the present level of atmospheric carbon dioxide through reducing deforestation, increasing afforestation or reforestation, and preventing biodiversity loss is a significant concern among scientists and policy makers (Kanowski et al., 2011; Pandey et al., 2014). The importance of engaging in meaningful action to combat deforestation is recognized in the United Nations Framework Convention on Climate Change (UNFCCC) and parties are discussing policies and approaches to reduce CO2 emissions from deforestation in a post-2012 international agreement on climate change. The UNFCCC recognizes various mitigation and adaptation options: firstly, the Clean Development Mechanism (CDM); secondly, Reduced Emissions from Deforestation and Forest Degradation REDD; and most recently the new strategy—reducing emissions from deforestation and forest degradation, and enhancing forest carbon stocks in developing countries (REDD+). These are intended to engage multi-scale stakeholders in conservation and sustainable management of forest resources for enhancing carbon sequestration in developing countries with incentives as a reward for mitigating global climate change (Gardner et al., 2012).

Afforestation and reforestation are the integral parts of REDD+ as an effective mechanism for reducing global climate change (Bonan, 2008; Wang et al., 2011; Pandey et al., 2014). The parties involved need accurate information on carbon stocks, biodiversity and the socioeconomic status of the communities in developing countries participating in the REDD+ financial mechanism (Pandey et al., 2014). Thus, protecting forest under REDD+ could be an effective measure to mitigate greenhouse gas emissions and will provide other ecosystem services, such as watershed management and biodiversity conservation (Pandey et al., 2014).

Our recorded 36 tree species in 34 genera and 17 families from roadside plantations was higher than recorded (19 species) from roadside plantations in Kanchanpur District, Nepal (Baral et al., 2013), but fewer than recorded (62 species) from a national highway plantation in Taiwan (Wang, 2011), or in homestead plantations in Southwestern Bangladesh (146 species; Kabir and Webb, 2008; 58 species; Motiur et al., 2006). These differences may reflect geographic and physiographic coverage, environmental gradients and the purpose of plantation management. For example, in roadside plantations fast-growing timber species are planted with the objective of getting a quick return, while in homegardens, multipurpose tree species, including fruit, timber, and ornamental species are planted.

Repeated disturbances such as grazing, soil work and others, favour species that grow best in degraded conditions, such as legumes (Brakenhielm and Liu, 1998). Indeed, the tree communities were dominated by several species of legumes in our study. Samanea saman (Rain tree), Dalbergia sissoo (Sissoo), Acacia nilotica (Babul), and Leucaena leucocephala (Ipil-ipil) had similar contributions to stem density, basal area, and biomass carbon (Table 1). Legumes accounted for 28% of the tree species and 94% of the biomass carbon in the five plantations (Table 2). Legumes also dominate in other forests in South Asia, including an urban forest of Chennai, India (Muthulingam and Thangavel, 2012) and natural forests in Tamil Nadu, India (Gamble and Fischer, 1934).

The stem density (3668 tree ha-1) in our study was much higher than recorded from other Asian countries. For example, 705 tree ha-1 was recorded from Taiwanese highway plantations of the same age (Wang, 2011) and 279 tree ha-1 from urban roadside forests in Shenyang, China (Liu and Li, 2012). Compared with some natural and restored forest systems in Bangladesh, our result for stem density and basal area (52.6 m2 ha-1) was higher than 381 trees ha-1 and 53.5 m2 ha-1 in Chittagong Hill Tracts (South) Forest Division (Nath et al., 1998), 459 trees ha-1 and 16.88 m2 ha-1 in Chunati Wildlife Sanctuary, Cox's Bazar (Rahman and Hossain, 2003), 464 trees ha-1 and 27.07 m2 ha-1 in Dudpukuria-Dhopachori Wildlife Sanctuary of Chittagong South Forest Division (Hossain et al., 2013), 257 tree ha-1 in Ukhiya natural forests of Cox's Bazar Forest Division (Ahmed and Haque, 1993) and 369 stem ha-1 in Bamu reserve forests of Cox's Bazar. This very high density of trees in roadside plantations compared to many other natural and restored forest systems in Bangladesh may be because of maintaining a certain tree spacing (1.5 m x 1.5 m) along with the higher survival rate from confirmed protection. Planting hedgerow crops, mainly Cajanus cajan on the inner side (close to the road) and Acacia nilotica on the outer side (away from the road), of roadside plantation is another reason for having higher stem density. Roadside plantations in southwestern Bangladesh may therefore play an important role in producing more timber and revenue for the local surrounding community as livelihood - and more importantly sequester carbon - compared to other natural and restored ecosystems in Bangladesh.

The niche complementary effect states that ecosystem functions like productivity and thereby carbon storage capacity are regulated by species richness and diversity (Tilman et al., 1996, 2006; Hooper et al., 2005)). However, we found a very weak insignificant positive relation between aboveground, belowground, and biomass carbon content with species diversity and evenness. Our finding is analogous to the study conducted in forest, agroforest and pasture land use system in tropical Panama (Kirby and Potvin, 2007), in Mexican tropical forest (Martinez-Sanchez and Cabrales, 2012), and in a mixed-species plantation in Sardinilla, Panama (Ruiz-Jaen and Potvin, 2011).

Our established three basal area based allometric models can be useful for carbon calculation from the plot level mean basal area as all three models showed strong relationships in the GLRM analysis (R2 = 0.84). The mean biomass carbon (192.80 Mg ha-1) was higher that the reported range (65-158 Mg ha-1) of tree biomass carbon for Bangladesh (Gibbs et al., 2007). Biomass carbon content was also higher than the reported average tree biomass carbon content of 83.72 Mg ha-1 (Shin et al., 2007) and 110.94 ton ha-1 (Ullah and Al-Amin, 2012) in hill forest of Bangladesh. Studies documented 11.71 ± 3.57 Mg ha-1 in Eastern Australia roadsides (Eldridge and Wilson, 2002), 45.49 Mg ha-1 in Buter Street and 22.29 Mg ha-1 in Penn Street, USA (Keating et al., 2005), 22.83 Mg ha-1 in USA national average urban forest carbon storage (Nowak and Crane, 2002), and 34.95 Mg ha-1 in roadsides of Shenyang, China (Liu and Li, 2012). We thus estimated a much higher amount of carbon in roadside plantations compared to other studies from tropical and subtropical regions. This may be due to the higher stem density and basal area (Table 2) in our study. This way roadside plantations in Bangladesh can play an important role in atmospheric carbon sequestration.

Despite their relatively poor physical condition and isolation, roadside plantations are still important for both livelihoods and carbon in Bangladesh. However, it is important to consider the role of the local village community in plantation establishment, protection, management, and conservation as they sustainably harvest non-timber products from the plantation to supplement livelihood. The villages would take the initiative to protect the plantation from further degradation, establish management rules, and enforce them with the active involvement of local forest officials. It is now widely acknowledged by social scientists that collective action and locally-initiated resources conservation can lead to successful conservation outcomes (Ostrom, 1990; Webb and Shivakoti, 2007). In addition, self-initiated protection, given rights to use, and strong community leadership contributed to safeguarding the plantations. Although natural and protected forests are usually more effective repositories of plant diversity than plantations managed for extractive use, participatory-managed plantations can be vital repositories of livelihoods and carbon alongside the protected area system in Bangladesh. This should be considered as a near-term participatory-managed conservation success, even if the level of diversity, livelihood supplement and amount of carbon stored estimated are low compared to other forest systems in Bangladesh under more complete protection.

Roadside plantations in southwestern Bangladesh still retain enough structural complexity to give a good start to livelihood supplementation and carbon sequestration. Given the challenges to long-term repositories to livelihood and carbon stocking discussed above, action should be taken to assist in the sustainable retention of roadside plantations in Bangladesh.

We recommend two investments that should increase the probability of long-term roadside plantation retention in Bangladesh. First, roadside plantations need to be facilitated with manual enrichment planting. Well-planned enrichment planting of species would maintain maximum density, improve soil conditions and create microenvironments favourable for other taxa such as wildlife. Enrichment planting would require a major human resource investment by the communities as well as generous technical backstopping and distribution of appropriate seedlings by the Forest Department, but could potentially have significant and positive impacts on forest conditions. Second, there needs to be strict implementation of operational rules, strengthening of institutions for regular monitoring, and increased authority to implement sanctions against violators. Vested interest in the plantation and its benefits, combined with the capacity and authority to protect the plantation, should improve the outcomes of plantation management by the communities. If both plantation establishment and vigilant protection is initiated promptly, roadside plantations in Bangladesh could be on a trajectory towards significant livelihood supplementation and carbon sequestration in addressing the long lasting poverty and environmental issues in Bangladesh respectively. This was the REDD+ mechanism of poverty alleviation and climate change addressed simultaneously.

Care must be taken when interpreting our results. Although indeed in this case study the participatory management did result in the retention of diverse plants in roadside plantations, the future trajectory of biodiversity conservation, livelihood supplementation and carbon sequestration potential is not known. Moreover, this single case study is not necessarily representative of all participatory managed roadside plantations in Bangladesh. Accumulation of longitudinal case studies will be required before we may generalize results like ours. Participatory managed roadside plantation may vary greatly in biodiversity, intensity of use, and trends of environmental services over time and geographical space. Thus, more research is required on the impacts of community managed roadside plantation for livelihood options and carbon sequestration potential in Bangladesh, and in Asia. The results of such research would have important implications in generalizing the impacts of managed landscapes, with particular reference to roadside plantations, on global poverty alleviation and climate change adaptation and mitigation strategies.

5. Conclusion

The participatory managed roadside plantations in southwestern Bangladesh contribute to livelihood supplementation from the pre-determined share (40%) of the receipts from the final harvest. Despite the relatively poor physical condition of the roadside plantations and their isolation, they are still important for livelihoods and for carbon sequestration. Although the plantations are harvested every 10-12 years, new plantations are established immediately after harvest, starting another cycle of carbon sequestration. The large area (4.65 million ha) of these plantations in Bangladesh suggests that their participation in the UNFCCC's financial based carbon mitigation strategies (e.g. CDM) could provide additional benefits to the local communities in a comanagement system such as Payment for Environmental Services (PES). Roadside plantations should be considered as a near-term participatory management conservation success, even if the contribution to livelihoods and the carbon pool estimated are low compared to other protected forest systems in Bangladesh. Their ability to contribute to livelihoods and carbon sequestration could be increased by management actions (enrichment planting and strict implementation of rules in use) to assist their sustainable retention. Despite this pioneering study, more research is needed into both the livelihood options and carbon sequestration potential of roadside plantations in Bangladesh and elsewhere in Asia.

Acknowledgements

The Working Plan Division of the Bangladesh Forest Department, Khulna, Bangladesh, provided financial and technical support for completing this research. Participants of the roadside plantations in the study area consented to data collection from their allocated areas and also provided help in fieldwork from time to time. The study area map was developed with direct assistance from Mustafizur Rahman and Mustafa Kamal Nice, and with the technical support from the Centre for Environmental and Geographic Information Services, Dhaka, Bangladesh. The Center for Southeast Asian Studies of Kyoto University, Japan, provided laboratory and financial support to Md. Enamul Kabir in finalizing the manuscript. The manuscript benefited from the comments of an anonymous reviewer.

References

Aguaron, E., McPherson, E.G., 2012. Comparison of methods for estimating carbon dioxide storage by Sacramento's urban forest. In: Lal, R., Augustin, B. (Eds.), Carbon Sequestration in Urban Ecosystems. Urban Ecosystems and Social Dynamics Program. USDA Forest Service, Research Park, California, USA, pp. 43-71.

Ahmed, G.U., Haque, S.M.S., 1993. Percentage distribution of species and diameter class in natural forest of Bangladesh. Univ. Stud. II 17 (1), 109-113. Albrecht, A., Kandji, S.T., 2003. Carbon sequestration in tropical agroforestry systems. Agricult. Ecosys. Environ. 99 (1-3), 15-27.

Baral, S.K., Malla, R., Khanal, S., Shakya, R., 2013. Trees on farms: diversity, carbon pool and contribution to rural livelihoods in Kanchanpur district of Nepal. BankoJanakari 23 (1), 1-63.

BBS, 2013. Statistical Year Book of Bangladesh. Bangladesh Bureau of Statistics (BSS), Statistic Division, Ministry of Planning, Bangladesh Secretariat. The

Government of the People's Republic of Bangladesh, Dhaka, Bangladesh. Bonan, G.B., 2008. Forests and climate change: forcings, feedbacks, and the climate benefits of forests. Science 320 (5882), 1444-1449. Brakenhielm, S., Liu, Q., 1998. Long-term effects of clear-felling on vegetation dynamics and species diversity in a boreal pine forest. Biodivers. Conserv. 7 (2), 207-220.

Cairns, M.A., Brown, S., Helmer, E.H., Baumgardner, G.A., 1997. Root biomass allocation in the world's upland forests. Oecologia 111(1), 1-11.

Carsan, S., Orwa, C., Harwood, C., Kindt, R., Stroebel, A., Neufeldt, H., Jamnadass, R., 2012. African Wood Density Database. World Agroforestry Centre, Nairobi, Kenya.

Chave, J., Andalo, C., Brown, S., Cairns, M.A., Chambers, J.Q., Eamus, D., Folster, H., Fromard, F., Higuchi, N., Kira, T., Lescure, J.-P., Nelson, B.W., Ogawa, H., Puig, H., Riera, B., Yamakura, T., 2005. Tree allometry and improved estimation of carbon stocks and balance in tropical forests. Oecologia 145 (1), 87-99.

Chave, J., Coomes, D., Jansen, S., Lewis, S.L., Swenson, N.G., Zanne, A.E., 2009. Towards a worldwide wood economics spectrum. Ecol. Lett. 12 (4), 351-366.

Dixon, R.K., Brown, S., Houghton, R.A., Solomon, A.M., Trexler, M.C., Wisniewski, J., 1994b. Carbon pools and flux of global forest ecosystems. Science 263 (5144), 185-190.

Dixon, R.K., Winjum, J.K., Andrasko, K.J., Lee, J.J., Schroeder, P.E., 1994a. Integrated land-use systems: assessment of promising agroforest and alternative land-use practices to enhance carbon conservation and sequestration. Clim. Change 27 (1), 71-92.

Eldridge, D.J., Wilson, B.R., 2002. Carbon storage in soil and vegetation in paired roadside sites in the box woodlands of eastern Australia. Aust. For. 65 (4), 268-272.

FAO, 2010. Forest Resources Assessment 2010. Food and Agriculture Organization of the United Nations, Rome, Italy.

Gamble, J.S., Fischer, C.E.C., 1934. Flora of the Presidency of Madras. Royal Botanic Gardens, Kew, London, p. 179.

Gardner, T.A., Burgess, N.D., Aguilar-Amuchastegui, N., Barlow, J., Berenguer, E., Clements, T., Danielsen, F., Ferreira, J., Foden, W., Kapos, V., Khan, S.M., Leesm, A.C., Parry, L., Roman-Cuesta, R.M., Schmitt, C.B., Strange, N., Theilade, I., Vieiram, I.C.G., 2012. A framework for integrating biodiversity concerns into national REDD+ programmes. Biol. Conserv. 154,61-71.

Gibbs, H.K, Brown, S., Niles, J.O., Foley, J.A., 2007. Monitoring and estimating tropical forest carbon stocks: making REDD a reality. Environ. Res. Lett. 2 (4), 1-13.

Hooper, D.U., Chapin, F.S., Ewel, J.J., Hector, A., Inchausti, P., Lavorel, S., Lawton, J.H., Lodge, D.M., Loreau, M., Naeem, S., Schmid, B., Setala, H., Symstad, A.J., Vandermeer, J., Wardle, D.A., 2005. Effects of biodiversity on ecosystem functioning: a consensus of current knowledge. Ecol. Monogr. 75 (1), 3-35.

Hossain, M.A., Hossain, M.K., Salam, M.A., Rahman, S., 2013. Composition and diversity of tree species in dudhpukuria-dhopachori wildlife sanctuary of Chittagong (South) forest division, Bangladesh. Res. J. Pharm. Biol. Chem. Sci. 4(2), 1447-1457.

IPCC, 2007. Fourth IPCC Assessment Report: Climate Change 2007. Cambridge University Press, Cambridge.

IPCC, 2013. Summary for Policymakers. In: Stocker, T.F., Qin, D., Plattner, G.K., Tignor, M., Allen, S.K., Boschung, J., Nauels, A., Xia, Y., Bex, V., Midgley, P.M. (Eds.), Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom, pp. 3-32.

Jashimuddin, M., Inoue, M., 2012. Management of village common forests in the Chittagong hill tracts of Bangladesh: historical background and current issues in terms of sustainability. Open J. For. 2 (3), 121-137.

Jose, S., 2009. Agroforestry for ecosystem services and Environmental benefits: an overview. Agrofor. Syst. 76 (1), 1-10.

Kabir, M.E., Webb, E.L., 2005. Productivity and suitability analysis of social forestry woodlot species in Dhaka Forest Division, Bangladesh. For. Ecol. Manag. 212(1-3), 243-252.

Kabir, M.E., Webb, E.L., 2008. Can home gardens conserve biodiversity in Bangladesh? Biotropica 40 (1), 95-103.

Kanowski, P., McDermott, C., Cashore, B., 2011. Post-Copenhagen strategies for the implementation of REDD+. In: Richardson, K., Steffen, W., Liverman, D., et al. (Eds.), Climate Change: Global Risks, Challenges and Decisions. Cambridge University Press, New York USA, pp. 429-430.

Keating, B., Roveda, E., Smith, M., Klemow, K., Toothill, W., Troy, M., 2005. Inventorying and Assessing the Values of Urban Trees in Kingston, PA using CITY green®5.0. Biology Department. Wilkes University, Wilkes-Barre, PA, p. 18766.

Kirby, K.R., Potvin, C., 2007. Variation in carbon storage among tree species: implications for the management of a small scale carbon sink project. For. Ecol. Manag. 246 (2-3), 208-221.

Kumar, B.M., 2011. Species richness and aboveground carbon stock in the home gardens of central Kerala, India. Agricult. Ecosys. Environ. 140 (3-4), 430-440.

Kumar, B.M., Nair, P.K.R. (Eds.), 2011. Carbon Sequestration Potential of Agroforestry Systems: Opportunities and Challenges. Springer.

Liu, C., Li, X., 2012. Carbon storage and sequestration by urban forests in Shenyang, China. Urban For. Urban Green. 11 (2), 121-128.

Ludwig, J.A., Reynolds, J.F., 1988. Statistical Ecology. John Wiley and Sons, New York.

Margalef, R., 1958. Information theory in ecology. Gen. Syst. 3, 36-71.

Martinez-Sanchez, J.L., Cabrales, L.C., 2012. Is there a relationship between floristic diversity and carbon stocks in tropical vegetation in Mexico? Afr. J. Agric. Res. 7 (17), 2584-2591.

Motiur, R.M., Furukava, Y., Kawata, I., Rahman, M., Alam, M., 2006. Role of homestead forest in household economy and factors affecting forest production: a case study in southwest Bangladesh. J. For. Res. 11 (2), 89-97.

Muthulingam, U., Thangavel, S., 2012. Density, diversity and richness of woody plants in urban green spaces: a case study in Chennai metropolitan city. Urban For. Urban Green. 11 (4), 450-459.

Nair, P.K.R., 2012. Carbon sequestration studies in agroforestry systems: a reality check. Agrofor. Syst. 86 (2), 243-253.

Nair, P.K.R., Nair, V.D., Kumar, B.M., Haile, S.G., 2009. Soil carbon sequestration in tropical agroforestry systems: a feasibility appraisal. Environ. Sci. Policy 12(8), 1099-1111.

Nath, T.K., Hossain, M.K., Alam, M.K., 1998. Diversity and composition of trees in Sitapahar forest reserve of Chittagong Hill Tracts (South) forest division Bangladesh.. Ann. For. 6(1), 1-9.

Nowak, D.J., Crane, D.E., 2002. Carbon storage and sequestration by urban trees in the USA. Environ. Pollut. 116 (3), 381-389.

Ostrom, E., 1990. Governing the Commons: The Evolution of Institutions for Collective Action. Cambridge University Press, New York, p. 280.

Pandey, S.S., Cockfield, G., Maraseni, T.N., 2014. Carbon stock dynamics in different vegetation dominated community forests under REDD+: a case from Nepal. For. Ecol. Manag. 327, 40-47.

Pielou, E.C., 1977. Mathematical Ecology. Wiley, New York, p. 385.

Rahman, M.L., Hossain, M.K., 2003. Status of fodder and non-fodder tree species in Chunati widlife sanctuary of Chittagong forest division, Bangladesh. Int. J. Usuj. Manage. 4 (2), 9-14.

Rahman, M.M., Khan, M.N.I., Hoque, A.K.F., Ahmed, I., 2014. Carbon stork in the Sundarbans mangrove forest: spatial variations in vegetation types and salinity zones. Wetl. Ecol. Manag. Online Early.

Roshetko, J.M., Lasco, R.D., Angeles, M.D., 2007. Smallholder agroforestry systems for carbon storage. Mitig. Adapt. Strateg. Glob. Change 12(2), 219-242.

Ruiz-Jaen, M.C., Potvin, C., 2011. Can we predict carbon stocks in tropical ecosystems from tree diversity? Comparing species and functional diversity in a plantation and a natural forest. New Phytol. 189 (4), 978-987.

Saha, S.K., Nair, P.K.R., Nair, V.D., Kumar, B.M., 2010. Carbon storage in relation to soil size-fractions undertropical tree-based land-use systems. Plant Soil 328(1-2), 433-446.

Shannon, C.E., Weaver, W., 1949. A Mathematical Model of Communication. University of Illinois Press, Urbana, IL.

Shin, M.Y., Miah, M.D., Lee, K.H., 2007. Potential contribution of the forestry sector in Bangladesh to carbon sequestration. J. Environ. Manag. 82(2), 260-276.

Takimoto, A., Nair, P.K.R., Nair, V.D., 2008. Carbon stock and sequestration potential of traditional and improved agroforestry systems in the West African Sahel. Agricult. Ecosys. Environ. 125 (1-4), 159-166.

Tilman, D., Hill, J., Lehman, C., 2006. Carbon-negative biofuels from low-input high-diversity grassland biomass. Science 314 (5805), 1598-1600.

Tilman, D., Wedin, D., Knops, J., 1996. Productivity and sustainability influenced by biodiversity in grassland ecosystems. Nature 379, 718-720.

Torres, A.B., Lovett, J.C., 2012. Using basal area to estimate aboveground carbon stocks in forests: La Primavera Biosphere's Reserve, Mexico. Forestry 00, 1-15.

Ullah, M.R., Al-Amin, M., 2012. Above-and below-ground carbon stock estimation in a natural forest of Bangladesh. J. For. Sci. 58 (8), 372-379.

UNFCCC, 2007. Climate Change: Impacts, Vulnerabilities and Adaptation in Developing Countries. Climate Change Secretariat (UNFCCC), Martin-Luther-King-Strasse 8, 53175 Bonn, Germany.

van der Werf, G.R., Morton, D.C., DeFries, R.S., Olivier, J.G.J., Kasibhatla, P.S., Jackson, R.B., Collatz, G.J., Randerson, J.T., 2009. CO2 emissions from forest loss. Nat. Geosci. 2, 737-738.

Wang, Y.C., 2011. Carbon sequestration and foliar dust retention by woody plants in thegreenbelts along two major Taiwan highways. Ann. Appl. Biol. 159 (2), 244-251.

Wang, W., Lei, X., Ma, Z., Kneeshaw, D.D., Peng, C., 2011. Positive relationship between aboveground carbon stocks and structural diversity in spruce-dominated forest stands in New Brunswick, Canada. For. Sci. 57 (6), 506-515.

Webb, E.L., Shivakoti, G.P., 2007. Decentralization, Forests and Rural Communities: Policy Outcomes in South and Southeast Asia. Sage Publications, New Delhi, India, p. 328. Eds..

Zanne, A.E., Lopez-Gonzalez, G., Coomes, D.A., Ilic, J., Jansen, S., Lewis, S.L., Miller, R.B., Swenson, N.G., Wiemann, M.C., Chave, J., 2009. Data from: towards a worldwide wood economics spectrum. Dryad Digital Repository. Global Wood Density Database. Retrieved from: http://dx.doi.org/10.5061/dryad.234 (accessed on December 26, 2014).

Zhang, Y., Duan, B., Xian, J., Korpelainen, H., Li, C., 2011. Links between plant diversity, carbon stocks and environmental factors along a successional gradient in a subalpine coniferous forest in Southwest China. For. Ecol. Manag. 262 (3), 361-369.

Zhang, J., Zhao, H., Zhang, T., Zhao, X., Drake, S., 2005. Community succession along a chronosequence of vegetation restoration on sand dunes in Horqin Sandy Land. J. Arid Environ. 62 (4), 555-566.