Concentration of atmospheric CO₂ has accelerated upward during the past few decades. In the last decade, the average annual rate of CO₂ increase was 1.91 parts per million (ppm). This rate of increase was more than double, as compared to the first decade of CO₂ measurements at the Mauna Loa Observatory 1 . The implications of increased concentration of CO₂ for climate and health of the global environment are topics of intense scientific, social and political concern. In contrast to economic globalization, no country can be left out of environmental globalization, as its consequences will sooner or later reach all. The direct solution to the problem is to reduce CO₂ emission 2 . Forests absorb CO₂ from atmosphere, and store carbon in wood, leaves, litter, roots and soil all acting as "carbon sinks". Carbon is released back into the atmosphere when forests are cleared or burned. Forests acting as sinks are considered to moderate the global climate. Overall, the world's forest ecosystems are estimated to store more carbon than the entire atmosphere 3 .

Quantifying the substantial roles of forests as carbon stores, as sources of carbon emissions and as carbon sinks has become one of the keys to understanding and modifying the global carbon cycle. Thus, estimating carbon stock in biomass is the most critical step in quantifying carbon stocks and fluxes from the forests. Hence the focus of this paper is to estimate the carbon stocks in India's forest biomass, taking into account the inventory data for diversified forest types present in the country. Numerous ecological studies have been conducted to assess carbon stocks based on carbon density of vegetation and soils 4 5 6 . The results of these studies are not uniform and have wide variations and uncertainties probably due to aggregation of spatial and temporal heterogeneity and adaptation of different methodologies. IPCC 7 estimated an average carbon stock of 86 tonnes per hectare in the vegetation of the world's forests for the mid-1990s. The corresponding carbon in biomass and dead wood in forests was reported 8 to be 82 tonnes per hectare for the year 1990 and 81 tonnes per hectare for the year 2005.

India is a large developing country known for its diverse forest ecosystems and biodiversity. It ranks 10th amongst the most forested nations of the world 3 with 23.84 percent (78.37 million ha) of its geographical area under forest and tree cover 9 . With nearly 173,000 villages classified as forest fringe villages, there is obviously a large dependence of communities on forest resources. Thus, it is important to assess the likely impacts of emission from forests on climate change, to develop and implement adaptation strategies both for biodiversity conservation and protection and for safeguarding the livelihoods of forest dependent people 10 .

Despite the importance of avoiding deforestation and associated emissions, developing countries have had few economic or policy incentives to reduce emissions from land use change 11 12 . 'Avoided deforestation' projects were excluded from the 2008-2012 first commitment period of the Kyoto Protocol because of concerns about diluting fossil fuel reductions, sovereignty and methods to measure emission reductions 13 14 . More recently the importance of including emissions reductions from tropical deforestation in future climate change policy has grown. The United Nations Framework Convention on Climate Change recently agreed to study and consider a new initiative, led by forest-rich developing countries, that calls for economic incentives to help facilitate reductions in emissions from deforestation and degradation in developing countries (REDD). The REDD has now become REDD plus and the core of REDD+ concept is to provide financial incentives to help developing countries voluntarily reduce national deforestation rates and associated carbon emissions below a baseline (based either on a historical reference case or future projection 10 .

The estimated physiographic zone wise carbon stocks in biomass of Indian forests are as follows. The various factors used in this study are given in Table 1.

The significance of forest area as a single indicator of forest development has often been overemphasized-growing stock and carbon storage may be considered equally important parameters. The net exchange of carbon between terrestrial ecosystems and the atmosphere remains one of the most uncertain components of global carbon budget. The studies on net carbon release from Indian forests due to land use changes, shifting cultivation etc., have come up with divergent results. Of the fourteen physiographic zones of country, only Western Himalayas, East Deccan, and Western Ghats have consistently shown an incremental change in carbon stock in both the assessment periods. The continuous increase in the western Himalayas may be attributed due to dense vegetation, and less disturbance as these forests are located on mountains with low population density, whereas increase in carbon stock in East Deccan may be due to increase in forest cover which has increased from 128,006 km² in 2003 to 128,757 km² in 2007. The maximum CO₂ sequestration was also found in this zone with an annual rate of 41.89 Mt CO₂ yr^-1 in ASP I and 54.55 Mt CO₂ yr^-1 in ASP II (Table 6).

Seven physiographic zones EH, NE, NP, EP, SD, WP, EG have shown negative change in carbon stocks, The maximum CO₂ emissions was found in EG zone with an annual rate of 70.22 Mt CO₂ yr^-1 in ASP I and 63.30 Mt CO₂ yr^-1 in ASP II (Table 6). The continuous decrease in the carbon stock in these five zones is attributed due to rapid urbanization and industrialization, as these forests are in plains having high population density.

The decrease in carbon stock in North East zone is mainly due to shifting cultivation. Shifting cultivation practice has cleared 0.05 Mha of forest area every year in northeastern states of India and total 17.22 Mt wood biomass and 10.69 Mt C was removed at the rate of 1.72 Mt and 1.07 MtCyr^-1 respectively 15 . The old age practice of shifting cultivation has been a single responsible factor for the forest and land degradation, there by changing the land use pattern. Around 0.45 million families in north east region annually cultivate 10,000 km² forests where as the total forests area affect by jhumming is believed to be 44,000 km² 16 .

The total carbon stock stored in forest biomass was 3325.30, 3223.96 and 3161.71 Mt with carbon density 42.93, 41.89 and 41.08 Ct ha^-1 for 2003, 2005 and 2009 respectively (Table 7). The total estimated carbon stock (wood only) reported by Manhas et al. 15 was 1085.06 Mt (or 1.09 Pg) and 1083.69 Mt (or 1.08 Pg) at a density of 24.94 and 24.54 t C ha^-1 for 1984 and 1994. Our estimated values are much higher than the values given earlier, which may be because these authors did not use the biomass expansion factor and did not consider the forest floor vegetation or it may be due increment in biomass due to increasing age structure of these forests. Kiswan et al. 10 have reported 2865.739 Mt of carbon stock in India's forest biomass for the year 2005 which is less than the present value (3223.96 Mt) calculated for 2005, which is due to the fact that these authors have considered 40% carbon in total biomass of country while we have considered 45% carbon in the total biomass. The carbon density t ha^-1 is within the ranged values given by Singh et al. 17 in their study of Central Himalayan forests. The carbon stock per unit area for Asian forests is 135, 90 and 40 t C ha^-1 (average, 88 t C ha^-1) for moist, seasonal and open forests respectively 18 , derived from wood volumes, and 250, 150 and 60 t C ha^-1 (average, 153 t C ha^-1) for moist, seasonal and open forests respectively 4 18 based on direct measurements.

The carbon stock in India's forest biomass decreased continuously from 2003 onwards despite slight increase in forest cover of the country from 23.57% in 2003 to 23.84% in 2007 (Table 7). The reason for decreasing carbon stock may be due to substantial loss of high density growing stock forests, while new forest plantations or regenerated areas have low carbon densities, which results in increase in forest area and decrease in carbon stock. According to Richards and Flint 19 , carbon storage in the vegetation in India from the year 1880 onwards shows a decreasing trend. The rate of carbon loss from the forest biomass in the ASP II had dropped by 38.27% in comparison to ASP I. This is mainly because the forest cover which earlier decreased from 23.57% in 2003 to 23.41% in 2005 thereafter increased to a record forest cover of 23.84% in 2007(Table 7). With the increase in forest cover the annual rate of CO₂ emission had dropped from 185.99 Mt yr^-1 in ASP I to 114.21 Mt yr^-1 in ASP II (Table 8).

The total estimated CO₂ loss from Indian forest biomass was 574.19 Mt at a rate of 185.99 Mt CO₂ yr^-1 for ASP I and 541.66 Mt at rate 114.21 Mt CO₂ yr^-1 for ASP II (Table 8). Each study in this connection has adopted a different approach based on different sources of data, different C pools for different years, resulting in net C flux that ranges from 0.4 Tg C yr^-1 20 to a sink value of 5 Tg C yr^-1 21 . The calculated carbon loss for Indian forests (wood only), reported by Manhas et al. 15 was 24.81 Mt C at a rate of 2.8 Mt C yr^-1 and 11.5 t C ha^-1 for the period of 1984 to 1994. Ravindranath et al. 21 reported that a total of 27.6 Mt C is emitted from the Indian forests annually as a result of deforestation and 12.87 Mt C from degraded forests. Houghton et al. 22 reported that forests hold more carbon per unit area in vegetation and soils than any other ecosystem that replaces them therefore conversion of forests into another land use also accompanies loss of biomass and carbon. Conversion of tropical forests to permanent agriculture and grazing lands has reduced the carbon density by 40%, whereas conversion to pasture has reduced the carbon content by 20% 23 . Houghton et al. 22 have reported that for 1980, approximately 80% of the net carbon flux from biota (2.0-2.5 Gt C yr^-1) is associated with change in land use in the tropics. Defries et al. 24 on carbon emission from tropical deforestation and re-growth, based on satellite observation for the 1980s and 1990s, noted that for the 1990s total carbon flux from tropical deforestation and re-growth is 0.95 Gt C yr^-1. The mean annual net C flux due to land use changes from Indian forests during 1880-1996 was estimated as 47 Tg C yr^-1. The cumulative net carbon flux from Indian forests due to land use change (deforestation, afforestation, and phytomass degradation) was estimated as 5.45 Pg C. Dadhwal et al. 25 had estimated the long-term carbon emissions of 3.45 Pg C from fossil fuel use (coal, lignite, petroleum and natural gas) and industrial activity (cement manufacture) for India during the 20th century. The mean annual net C flux for the recent period (1985-1996) due to land use changes was estimated as 9.0 Tg C yr^-1. This was an attempt to study the long-term pattern of net carbon flux from Indian forests with regional variabilities also, which has not been reported in earlier studies. The major decline in forest area could be accounted for a high net carbon release from Indian forests before the 1980 period. As a result of various afforestation programmes, forest conservation efforts, and joint forest management programmes by various government and non-government agencies, the forest area of India has stabilized to approximately 64 Mha, during the recent period, i.e., after 1980 26 . The increase in forest area resulted in little uptake of carbon in Indian forests. For the recent period the Indian forests are a small source of carbon, which compares favorably with peak net annual carbon emissions during 1970-1980s 26 . The rate of afforestation in India is 2 Mha per annum, which is considered to be one of the highest among the tropical countries 27 .

The Bali Action Plan included all the essentials of forest improvement i.e. reducing deforestation, conservation and sustainable management and enhancement of forest carbon stocks. Whether forests act as reservoirs, sinks for carbon from the atmosphere, or sources of GHGs depends on several factors such as the age of the forest, the management regime, other biotic and abiotic disturbances (e.g. insect pests, forest fires, etc.) and human-induced deforestation. Planting forests (afforestation and reforestation) clearly provides an opportunity to sequester carbon in vegetation and soils. However, it takes decades to restore carbon stocks that have been lost as a result of land-use changes. The reduction of deforestation and enhancement of forest carbon stocks are the two sides of the same coin, where one cannot do without the other. Both are equally important.

In the present study an attempt was made to estimate the carbon stock of the country for two assessment periods, because estimating forest biomass carbon is the most critical step in quantifying carbon stocks and fluxes from forests. The carbon stock in India's forest biomass decreased continuously from 2003 onwards, despite slight increase in forest cover. Increasing forest cover will not help in REDD implementation unless deforestation and degradation will not be reduced because country's forest cover has already been degraded and dense forests are losing their crown density and productivity continuously. REDD implementation will be a challenge for India because of the complexity of the different elements influencing deforestation and forest degradation and requires a range of policy approaches and positive incentives to address the challenges. However with the rate of carbon loss from the forest biomass in the ASP II had dropped by 38.27% in comparison to ASP I. This is mainly because the forest cover which earlier decreased from 23.57% in 2003 to 23.41% in 2005 thereafter increased to a record forest cover of 23.84% in 2007 It was observed that with the increase in forest cover of the country the CO₂ emission from the forestry sector had also started decreasing. With the Copenhagen Accord, India along with other BASIC countries China, Brazil and South Africa is voluntarily going to cut emissions. India will voluntary reduce the emission intensity of its GDP by 20-25% by 2020 in comparison to 2005 level. Activities like REDD+ can provide a relatively cost-effective way of offsetting emissions, either by increasing the removals of greenhouse gases from the atmosphere by afforestation programmes, managing forests, or by reducing emissions through deforestation and degradation.

The total standing above-ground biomass of woody vegetation is often one of the largest carbon pools. The above-ground biomass comprises all woody stems, branches, leaves of living trees, creepers, climbers, and epiphytes as well as herbaceous undergrowth. Estimation of carbon stocks stored in Indian forests, in the present study is based on the secondary data of growing stock data published by Forest survey of India 28 29 9 in the State of Forest Reports. Assessment of biomass was based on the consideration that all lands, more than one hectare in area, with a tree canopy density of more than 10 per cent are defined as 'Forest'. The satellite data used for 2003 related to the period from 2001 to 2003, the data for the 2005 assessment related to the period from 2003 to 2005, and for 2009, pertained to 2006 to 2007.

Suitable biomass increment values (expansion and conversion for calculating total tree above ground biomass) and the ratio of below and above ground biomass (for calculating total tree biomass above and below ground) as available in different studies covering a range of forest types of the country were used in the present study. The various factors used in this study are given in Table 1.

Estimation of carbon stocks present in biomass is based either on IPCC (Good Practice guidelines (IPCC, GPG, 2003) or published literature for conversion and other factors starting from the growing stock (GS) data of forest inventories. The biomass in this study was calculated as;

AGB = G stk × MD × B exf

Where,

AGB Above Ground Biomass Mt

G_stk = Growing Stock in Mm³

MD = Mean density

B_exf = Biomass expansion factor

The Below Ground Biomass was calculated by root shoot ratio:

BGB = AGB × R bel . ab

Where,

BGB = Below Ground Biomass Mt

R_bel.ab = Ratio Below to Above Ground Biomass

The total biomass was estimated as;

T bm = AGB + BGB

In general, other forest floor biomass accounts for less than 2 percent of total biomass of closed forest formations 30 31 . However for this study, ratio was adopted based on the published records for different vegetation types and different localities, and also keeping in view its application and representation for the country level estimates 32 33 34 35 . The forest floor biomass was estimated by the following.

F fb = T bm × R tbm

Where,

F_fb = forest floor biomass in Mt

T_bm = Total biomass in Mt

R_tbm = Ratio to total biomass in Mt

Total forest biomass was estimate as;

TFB = T bm + F fb

To estimate the total dry weight in biomass 80% of total forest biomass was considered. Biomass material contains about 40% carbon by weight. The variability of approximately 9% depends on the nature of the biomass material 36 37 although most studies have used the carbon proportions between 40 to 50% depending on the requirements 38 39 40 41 42 43 .

The carbon content of vegetation is surprisingly constant across a wide variety of species. Most of the information for carbon estimation described in the literature suggests that carbon constitutes between 45 to 50 percent of dry matter 44 45 . To estimate the total amount of carbon stocked in India's forests, dry weight of biomass was converted into carbon by multiplying with a factor of 0.45 as used by Woomer 46 .

The change in Carbon stocks was assessed by the stock change method as per IPCC guidelines.

Δ C = C 2 - C 1

ΔC = Change of carbon stock

C₂ = Carbon stock at time 2

C₁ = Carbon stock at time 1

For annual change in C stocks following equation were used.

Δ C = ( C 2 - C 1 ) ∕ ( t 2 - t 1 )

One ton of carbon in wood or forest biomass represents 3.67 tons of atmospheric carbon dioxide. Total atmospheric CO₂ accumulated or emitted by the forest biomass was estimated by multiplying the carbon Stock values with 3.67, the molecular weight of Carbon dioxide. The total carbon stock change in India was estimated by Gain loss (default) method.

Δ C = Δ C G - Δ C L

The authors declare that they have no competing interests.

MAS, MK and NPT share the contributions to data analysis, and compilation of this manuscript. RWB shares his valuable contribution from initial manuscript drafting to final submission. All authors read and approved the final manuscript.