Phytosociological assessment and carbon stock estimation and valuation in the tropical dry deciduous forest of Bihar

Purpose – This study aims to assess the biodiversity of the study area and estimate the carbon stock of two dry deciduous forest rangesof BankaForest Division, Bihar, India. Design/methodology/approach – The phytosociological analysis was performed and C stock estimation based onvolumedetermination throughnondestructivemethods wasdone. Findings – Phytosociological analysis found total 18,888 [14,893 < 10 cm (diameter at breast height) dbh] and 2,855 (1,783 < 10 cm dbh) individuals at Banka and Bounsi range with basal area of 181,035.00 cm 2 and 32,743.76 cm 2 , respectively. Importance value index was highest for Shorea robusta in both the ranges. Species diversity index and dominance index, 1.89 and 1.017 at Banka and 1.99 and 5.600 at Bounsi indicated the prevalence of biotic pressure. Decreased dbh and tree height resulted in a lowered growing stock volume as 59,140.40cm 3 ha (cid:1) 1 (Banka) and 71,306.37 cm 3 ha (cid:1) 1 (Bounsi). Total C stock at Banka and Bounsi range was 51.8 t ha -1 and 12.56 t ha (cid:1) 1 , respectively where the highest C stock is recorded for Shorea robusta in both the ranges (9.8 t ha (cid:1) 1 and 2.54 t ha -1 , respectively). A positive correlation between volume, total biomass and basal area of tree species with C stock was observed. R 2 value for Banka range was 0.9269 (volume-C stock), 1 (total biomass-C stock) and 0.647 (basal area-C stock). Strong positive correlation was also established at Bounsi range with R 2 value of 1. Considering the total forest area enumerated, C sequestration potential was about 194.25 t CO 2 (Banka) and 45.9 t CO 2 (Bounsi). The valuation of C stock was therefore US$2,525.25 (Banka) andUS$596.70 (Bounsi). Practical implications – The research found the potentiality of the study area to sequester carbon. However, for future, the degraded areas would require intervention of management strategies for restoration of degradedlands andprotectionof planted treesto increase the carbonsequestrationpotentialof thearea. Originality/value – Present study is the ﬁ rst attempt to assess thephytosociology and estimate the regulatory servicesofforestwithrespecttobiomassandcarbonstockestimationfortheBankaforestdivisionofBihar.


Introduction
One of the richest terrestrial ecosystems is constituted in tropical forests supporting various life forms that indeed maintain high biodiversity (Shi and Singh, 2002). Eighty-six percent of the forest land is contributed by tropical forests in India while contribution of tropical dry deciduous forest and moist deciduous forest is 53% and 37%, respectively. Share of wet evergreen and semi-evergreen forests is only 10%. Tree species diversity is both complex and varies in different places in its structure and composition due to the prevalence of varying climate and topographical characteristics (Raturi, 2012). Depending upon the structure and composition of forests functionality is determined where forests act as carbon sink and have potential to sequester carbon (Lal and Singh, 2000). The phytosociological studies are significant to understand the structure, composition and distribution pattern of plant communities (Rout et al., 2018) and also to estimate the biomass of the area. Estimation of biomass eventually contributes to estimate C stock of an area (Fahey et al., 2010;Kushwaha et al., 2014;Salunkhe et al., 2016;Jhariya, 2017;Banik et al., 2018). Studies have depicted that carbon stocks are dependent on forest tree density, volume, above-and belowground biomass (Gibbs et al., 2007;Banik et al., 2018). The estimates of percentage indicate higher priority for tropical dry deciduous forests but limited studies have been conducted in these forests. Many of the forests are subjected to maltreatment and are degraded. Both biotic and edaphic factors have accelerated the process of degradation finally turning the rich dense forests into open degraded and scrub lands (Singh et al., 1991;Chaturvedi et al., 2011). Banka forest division is tropical dry deciduous forest with forest fringe villages and is under immense biotic pressure on forested land causing degradation of forest area leading to loss of biodiversity, habitat fragmentation, removal of top soil, etc. The loss of biodiversity in dry deciduous forest of tropics is comparable to tropical forests (Gentry, 1992).
Deforestation and land degradation causes loss of carbon stocks or in other words emits CO 2 , which estimates about 7%-14% of the total CO 2 emissions from anthropogenic activities (Harris et al., 2012;Achard et al., 2014). A decreasing trend in carbon stocks of tropical forests in India is noticed since 2003 (Sheikh et al., 2011) with reduction in native forests at the rate of 3.5% annually (Puyravaud et al., 2010). Similarly, the decrease in global forest area was noticed by 4.1 and 6.4 million ha annually and 3% of world's forest were disturbed by several biotic factors, namely, fire, pests, logging, etc. as reported by FAO (2012), while it was also reported by FAO (2006) that about 60% of forests are recovering.

Tropical dry deciduous forest of Bihar
Extensive studies were made by several researchers on the deforestation having impact on climate and what role is played by tropical forests in climate change mitigation (Masera et al., 1995;De Jong et al., 1999Grace et al., 2006). It was further estimated that 89% of total carbon stored in an ecosystem is lost due to deforestation that leads to loss of living biomass (Keith et al., 2014). However, United Nations Framework Convention on Climate Change was set and estimation of forest carbon sinks, as well as sources, was in demand to inventories (UNFCCC, 1992). Major sources of carbon sink are the forests and are, thus, required to assess the total amount of sequestered carbon. Higher priority for adaptation and mitigation of climate change issues was set for conservation and protection of biological diversity and carbon sequestration (Diaz et al., 2009). In recent past, under REDDþ programs for implementation of climate change mitigative policies, the developing countries are required to furnish baseline data for carbon stock estimation in forests (Saatchi et al., 2011;Salimon et al., 2011).
The estimated area for Sal forest in India is about 13 million hectares where in most cases the primary Sal forest is replaced by secondary regenerated Sal forest. The major cause for the shift was due to forest land degradation, over-exploitation, deforestation, grazing, change in land use pattern and several other biotic and anthropogenic activities (Deka et al., 2012). As the species diversity and composition is dependent on potential regeneration of secondary forests (Ayyappan and Parthasarthy, 1999), the biomass of forest is also modified that has direct impact on carbon storage. Thus, forests being largest pool of biomass and carbon, different percent coverage has been estimated for aboveground, belowground, dead woody and litter compartment, which is about 234 Pg C, 62 Pg C, 42 Pg C and 23 Pg C, respectively with soil carbon pool of 398 Pg C (Kindermann et al., 2008). Therefore, considering the carbon pool although many studies were carried out by many researcher but limited studies are done in tropical dry deciduous forest in eastern zone of India. Few studies on biomass and carbon estimation shows total carbon pools of 52.59 Mg ha À1 , 34.17 Mg ha À1 and 33.61 Mg ha À1 at Ailanthus excels -Cassia fistula forest, Acacia leucophloea -Balanites aegyptica forest and Anoegeissus pendula -Acacia leucophloea dominated forest, respectively, in North-east India . Biomass allometric equations were used by few researchers to estimate biomass and carbon stock in tropical forest of Tripura that recorded biomass in the range of 37.85 to 85.58 Mg ha À1 (Majumdar et al., 2016). Forest of Manipur showed carbon stock in the range of 60.09 to 121.43 t ha À1 (Thokchom and Yadava, 2017) Similar studies at Garhwal Himalaya, India recorded 132.74 and 66.36 Mg ha À1 of total biomass and carbon density, respectively (Mahato et al., 2016).
In view to the above issues, as vegetation structure and diversity plays a major role in controlling various ecological processes (Gower et al., 1992;Rout et al., 2018), our study was concentrated to Banka Forest Division of Bihar state where traditional process of forest vegetation survey was followed to assess the phytosociology and estimate the regulatory function with respect to biomass and carbon stock estimation in the study area. In our present study, forest structure, distribution and carbon stock is estimated with its valuation in two forest ranges, namely, Banka and Bounsi range of Banka Forest Division, Bihar.

Study area
The study was conducted in Banka and Bounsi Range with geographical location as latitude 24°30 0 00 00 N to 25°15 0 00 00 N, longitude 86°30 0 00 00 E to 87°15 0 00 00 E. The total area of Banka and Bounsi forest Range is 15,106.579 ha and 6,760.000 ha, respectively. The area receives rain during onset of southwest monsoon in the month of June scaling to 1,200 mm precipitation annually. EFCC 2,1 2.2 Sampling plots Sampling plots were selected based on geo-referenced toposheets of 1:50000 scale. The study area was cropped and 15 0 Â 15 0 grid was subdivided into 144 sub-grids. Each sub-grid of size 1.25 0 Â 1.25 0 was further subdivided into to nine sub-grids of 25 00 Â 25 00 using Arc GIS software.

Vegetation enumeration
Vegetation was enumerated following nested quadrat was laid in each sampling plots of 25 00 Â 25 00 covering 0.5 ha. The phytosociological analysis was done to assess the structure of vegetation following Misra (1968) where diameter at breast height (DBH) in cm and tree height (in m) was measured. Frequency, density and abundance were recorded (Curtis and McIntosh, 1950). Relative values were calculated following Philips (1959). The importance value index (IVI) was estimated as sum of relative frequency, relative density and relative dominance of each species (Curtis, 1959). Shannon Weiner index (H 0 ) was used to calculate species diversity (Shannon and Wiener, 1963). The equations are as follows: Frequency where n i is the total number of individuals of species i and N is the total number of individuals of all species.

Carbon stock estimation
Tree Basal Area TBA ð Þ was calculated based upon the formula Area A ð Þ ¼ p r 2 ; Volume (m 3 tree À1 ) of each tree in a sampling quadrat obtained is converted into the volume on hectare basis. Above ground biomass (AGB) was calculated following IPCC (2003). Below ground biomass (BGB) was calculated following the equation given by Mokany et al. (2006). The carbon storage for each species was computed by multiplying total biomass with constant factor 0.50 (IPCC, 2006).

Growing stock
As the forest is under successional stage and growing stock estimation reveals most of the trees (immature) fall under dbh class 0-10 cm, therefore total carbon stock estimation is determined for all dbh classes. The carbon stock estimation is limited to live tree biomass in our study. Height of trees in all dbh class ranges between 1.0-14 m and 1.5-15 m in Banka and Bounsi range, respectively. The total volume of all the trees enumerated is recorded as 59,140.40 and 71,306.37 cm 3 ha À1 in Banka and Bounsi range. Tree height and dbh is one of the major factors for contributing to large amount of stock volume. Although tree height for both the ranges are almost same, the number of trees above 10 cm dbh is more in Bounsi range (37%) than compared to Banka range (26 %). Tree density at Banka is higher than Bounsi but larger number of trees in higher dbh class has contributed to higher volume of growing stock at Bounsi range. In both the ranges, volume of Shorea robusta is higher than other species. At Banka range the top five tree species contribution to volume follows the pattern as Shorea robusta (9,520.00 cm 3 ha À1 ) > Madhuca indica (9,001.56 cm 3 /ha) > Mangifera indica (5,946.00 cm 3 ha À1 ) > Colchlospermum religiosum (4,492.38 cm 3 ha À1 ) > Acacia auriculiformis (4,199.68 cm 3 ha À1 ). Similarly, at Bounsi range the pattern is Eucalyptus globulus (32,893.80 cm 3 ha À1 ) > Shorea robusta (13,789.00 cm 3 ha À1 ) > Madhuca indica (6,024.11 cm 3 ha À1 ) > Butea monosperma (4,849.29 cm 3 ha À1 ) > Acacia auriculiformis (2,584.16 cm 3 ha À1 ) (Tables 3 and 4). The pattern of AGB for top five tree species at Banka range is Shorea robusta > Madhuca indica > Mangifera indica > Terminalia bellerica > Acacia auriculiformis. AGB of all the tree species at Banka and Bounsi is 2,196.29257 and 75.312 kg ha À1 , respectively. Likewise, BGB is recorded as 15.008 and 17.698 kg ha À1 , respectively for Banka and Bounsi range, therefore total biomass of two ranges are 78.875 and 93.011 kg ha À1 , respectively.
Total C stock at Banka and Bounsi range is 39.44 and 46.51 kg ha À1 , respectively where highest C stock is recorded for Shorea robusta in both the ranges (Banka -7.65 kg ha À1 ; Bounsi -9.40 kg ha À1 ) (Tables 3 and 4). Depending upon the growing stock and C stock, clustering of tree species was done and dendrogram (Figures 3 and 4) shows three clusters in both the ranges, which are different from the pattern for phytosociology. It was recorded that in Banka range, Madhuca indica and Shorea robusta formed Cluster 1 (Sapotaceae and Tropical dry deciduous forest of Bihar Dipterocarpaceae family, respectively) while Mangifera indica and Terminalia bellerica formed Cluster 2 (Anacardeaceae and Combretaceae family, respectively). Rest of the tree species have similar association and formed cluster 3 that includes 28 species belonging to 17 families (Anacardiaceae, Annonaceae, Apocyanaceae, Bixaceae, Combretacaea, Ebenaceae, Fabaceae, Lamiaceae, Meliaceae, Mimosaceae, Moraceae, Myrtaceae, Phyllanthaceae, Rhamnaceae, Rutaceae, Sapindaceae and Sapotaceae). Similarly, in Bounsi range, Eycalyptus globulus and Shorea robusta formed two clusters, namely, Clusters 1 and 2, respectively and rest of the tree species formed cluster 3 that includes 23 tree species and 14 families (Anacardiaceae, Bixaceae, Combretaceae, Ebenaceae, Fabaceae, Lamiaceae, Malvaceae, Meliaceae, Moraceae, Myrtaceae, Rhamnaceae, Rutaceae, Sapindaceae and Sapotaceae).

Forest structure
Abundance: frequency ratio recorded for Banka and Bounsi are 0.187 and 0.151, respectively depicting cluster distribution of trees (Figures 1 and 2). Although IVI is highest for Shorea Notes: SGspecific gravity (g cm À3 ); Vvolume (cm 3 ha À1 ); ABGabove ground biomass (g ha À1 ); BGBbelow ground biomass (g ha À1 ); TBtotal biomass (g ha À1 ); Ccarbon stock (kgC ha À1 ); CO 2 equivalent (kg CO 2 /ha) EFCC 2,1 robusta but the values are lower than the values recorded for Doon Valley at Western Himalaya (Gautam et al., 2008;Mandal and Joshi, 2014) and other tropical forests (Ganguli et al., 2016). The species diversity index value in our study is comparable to Eastern Ghats, which is much higher than the value in our study (Reddy et al., 2008;Ganguli et al., 2016) and consistent with studies made by Saklani et al. (2018). Correlation between biomass and volume of growing stock depicts dependency of total biomass on growing stock volume (Banka: R 2 = 0.9269, Bounsi: R 2 = 0.9943) where dbh and height are the predictor for volume of tree individuals. Positive correlation was also observed between volume, biomass, basal area of tree species with carbon stock (Banka range: R 2 = 0.9269 for Volume-C stock; R 2 = 1 for total biomass-C stock and R 2 = 0.647 for basal area-C stock; and Bounsi range: R 2 = 1 for volume, biomass and basal area with C stock). However, sometimes specific gravity of trees also contributes to the biomass; therefore the pattern for total biomass differs from volume pattern of tree species at Banka range (specific gravity of Cochlospermum religiosum is 0.270 g cm À3 that caused the lowering of AGB than Acacia auriculiformis and Terminalia bellerica at Banka range). Similar studies on biomass and C storage at Sathanur Reserve Forest of eastern ghats and Asola Bhatti Sanctuary in Northern Aravalli hills showed positive correlation between biomass and C storage where highest storage was contributed by Albizia amara and Anogeissus pendula, respectively (Kushwaha et al., 2014;Salunkhe et al., 2016;Jhariya, 2017). The C stock in our study shows lowered value, which is comparable to the values estimated Notes: SGspecific gravity (g cm À3 ); Vvolume (cm 3 ha À1 ); ABG -above ground biomass (g ha À1 ); BGBbelow ground biomass (g ha À1 ); TBtotal biomass (g ha À1 ); Ccarbon stock (kg Cha À1 ); CO 2 equivalent (kg CO 2 /ha) Tropical dry deciduous forest of Bihar at Garhwal Himalaya (Mahato et al., 2016) and Manipur (Thokchom and Yadava, 2017) and Tripura (Banik et al., 2018) while the values are similar to the C stock estimated for North East India . Lowered biomass in our study is relatively due to presence of young trees having <10 cm dbh, small bole size and often anthropogenic disturbances such as lopping and grazing, prevailing in the area caused removal or lowering of biomass, which is similar with study at tropical deciduous forest of Madhya Pradesh, India (Salunkhe et al., 2016;Dar et al., 2019). However, Prevalence of edaphic factors with poor soil depth and soil structure is also responsible for low above ground biomass.

Carbon sequestration potential
With respect to the C stock recorded for all the trees enumerated in Banka and Bounsi range have potential to sequester 144.74 kg CO 2 ha À1 (or 0.15 t CO 2 ha À1 ) and 170.50 kg CO 2 ha À1 (or 0.17 t CO 2 ha À1 ), respectively. Therefore, extrapolating the amount of C stock and CO 2 sequestered in the total forest area enumerated it is about 51.8 t ha À1 and 194.25 t CO 2 at Banka range and 12.56 t ha À1 and 45.9 t CO 2 at Bounsi, respectively. The amount of C sequestered is much lower than compared to other tropical forests of North Western Ghats (Mandal and Joshi, 2014;Patel et al., 2015;Salunkhe et al., 2016;Banik et al., 2018;Dar et al., 2019). Lowered amount is contributed by lowered C stock, which indeed depends upon dbh and volume, which acts as an important indicator for C stock in trees. However, the dominant tree species Shorea robusta solely contributes to about 19% and 20% of total C stock of enumerated area at Banka and Bounsi forest range, respectively.

Conclusions
The study illustrates forest structure and pattern of distribution of trees in the area, which also determines the biomass and carbon stock pattern in the study area. From the survey, it is evident that most of the forest area is degraded and many of the area are restored through plantation and afforestation programs. Coppice Sal (Shorea robusta) is noticed in most of the forest area under two ranges and the forest undergoes its secondary successional stage supports the restoration of forests through afforestation. However, the prevalence of high biotic pressure in different pockets of the study area accompanied with edaphic factors causes reduced survival rate of recruited plants in terms of recruitment, growth and establishment. Lowered growing stock volume accompanied by lowered biomass is subjected to decreased C stock value compared to other forests of tropics, which is due to the presence of maximum trees under dbh class <10 cm. However, the valuation of these forests (area enumerated only) in terms of C sequestration (for present year of C stock estimation) applying international price @ US$13 t À1 CO 2 is estimated as US$2,525.25 and US$596.70 for Banka and Bounsi range, respectively. Therefore, these forests have potential to act as carbon sink but proper management and protection is required for young trees along with habitat restoration and biodiversity conservation of the entire forest ranges.