Ecological Indicators Spatial distribution of forest ...

Ecological Indicators 85 (2018) 742–752

Contents lists available at ScienceDirect

Ecological Indicators journal homepage: www.elsevier.com/locate/ecolind

Spatial distribution of forest biomass carbon (Above and below ground) in Indian forests

T

⁎

Gopalakrishnan Rajashekara, Rakesh Fararodaa, , R. Suraj Reddya, Chandra Shekhar Jhaa, K.N. Ganeshaiahb, Jamuna Sharan Singhc, Vinay Kumar Dadhwala a

National Remote Sensing Centre, Indian Space Research Organisation, Balanagar, Hyderabad, 500 037, India University of Agriculture Science, Bangalore, 560 065, India c Department of Botany, Banaras Hindu University, Varanasi, 221 005, India b

A R T I C L E I N F O

A B S T R A C T

Keywords: Forest inventory Remote sensing Growing stock Wood density Allometric equations Forest carbon density

Forest carbon (C) estimates are the key inputs to the understanding of the global C cycle. We report the estimates of forest carbon pool and its spatial distribution in the Indian forests for the years 1994 and 2010 at 5 km grid level. This study improves upon earlier spatial estimates of Indian forest biomass carbon by using data from a robustly designed National Forest Inventory (NFI). The realized sampling intensity has addressed the large heterogeneity of the Indian forest types and allowed the computation of 5 km grid level forest C, yielding a realistic estimate of forest biomass C in Indian forests. Forest cover density maps were intersected with 5 km mesh and estimates of forest area, forest carbon density for each Agro-ecological sub region and forest carbon pools were linked to the 5 km grid coverage of India. National forest carbon estimates for the years 1994 and 2010 are 3911.78 and 4368.03 TgC respectively, and these estimates showed a net increase of 456.25 TgC in 16 years. Uncertainty of the estimates has been addressed spatially. Mean forest carbon density increased from 61.14 Mg ha−1 in 1994 to 64.08 Mg ha−1 in 2010. C densities for dense and open forest in 1994 estimated as 77.08 and 38.47 Mg ha−1 with total C pools of 2895.28 TgC and 1016.50 TgC which has increased to 80.24 Mg ha−1 and 41.69 Mg ha−1 with total C pools of 3176.48 TgC and 1191.55 TgC in 2010. This study provides the first 5 km level C analysis for Indian forests. Spatial distribution of C shows large differences in C density over Indian forests indicating that estimates of the spatial distribution of C are even more important than the total C pool estimates of the country.

1. Introduction The increasing carbon dioxide (CO2) loading in the atmosphere has become one of the major global environmental issues in recent years because increasing atmospheric CO2 causes climate change (Stocker et al., 2013). Forests contain about 80% of global terrestrial aboveground carbon stocks in their vegetation and soils, and also exchange large quantities of carbon with the atmosphere through photosynthesis and respiration playing an important role in the global carbon (C) cycle (Noble et al., 2000). Estimations of forest biomass and its change are important for assessing historical and present anthropogenic C releases from forests and also in evaluating the possibilities of future potential C sequestration (Ciais et al., 2013; Tan et al., 2007). Forested areas can behave as a source of atmospheric carbon when they get disturbed by human or natural causes, and an atmospheric carbon sink during the regrowth after disturbance, and hence they can be managed to alter the

magnitude and direction of their C fluxes (Brown et al., 1996). Forest carbon inventories at national level and state/regional level are essential for developing countries, as well-authenticated estimates of forest carbon stocks are necessary for successful implementation of mitigating policies to take advantage of the REDD programme (Saatchi et al., 2011). Pan et al. (2013) has estimated the global forest C pool as 363 PgC, based on global aggregation of forest inventories and field observations. Analysis of forest C cycle indicates large spatial differences in C pools, fluxes and net C releases due to land use changes, environmental and anthropogenic factors (Chhabra and Dadhwal, 2004; Achard et al., 2004; Houghton, 2008). Improved understanding of forest C pools is important for the sustainable development and conservation of forests, formulating strategies for carbon sequestration, and accurate estimation of contribution of land use changes to C emission in India (Chhabra et al., 2002a).

Abbreviations: GS, growing stock; Mg ha−1, million gram per hectare; Mha, million hectare; TgC, tera gram carbon; WD, wood density; BEF, biomass expansion factor ⁎ Corresponding author. E-mail address: [email protected] (R. Fararoda). https://doi.org/10.1016/j.ecolind.2017.11.024 Received 30 September 2016; Received in revised form 31 August 2017; Accepted 13 November 2017 1470-160X/ © 2017 Elsevier Ltd. All rights reserved.


G. Rajashekar et al.

The carbon values reported in Table 1 are mostly based on the global vegetation maps generated earlier (Whittaker and Likens, 1973; Olson et al., 1983) as well as from regional studies (Houghton et al., 1995), and include both above and below ground and ground cover Carbon. C estimates by Brown (1997) were based on forest inventory data converted to C stock using allometric models, and Achard et al. (2004) increased these values by 20% for root carbon stocks. Estimates by Gibbs et al. (2007) are based on method pioneered by Brown et al. (1993) using a rule base GIS analysis to spatially extrapolate forest field inventory data based on climate, soils, topography, population and land use information and produce a map of forest carbon stocks in the 1980 (Iverson et al., 1994; Brown and Gaston, 1995). Gibbs and Brown (2007) estimated carbon values for South East Asia by taking the average forest carbon stock for each biome from maps that represent actual forest carbon in 2000. IPCC (2006) default values for forest biomass, mostly based on Penman et al. (2003), are based on compilation of published studies. Biomass values presented in IPCC (2006) converted to carbon stock using IPCC default carbon conversion factor (carbon fraction of biomass = 0.47), and below ground carbon stock added using the ratio of belowground biomass to the aboveground biomass.

This study presents the spatial forest carbon stock estimates at 5 km grid for the years 1994 and 2010. This study indicates that Indian forests are sequestering carbon, and carbon stored in Indian forest has increased over the study period. 1.1. Carbon (C) estimation of Indian forests –earlier studies India ranks fifth in the world in carbon dioxide emissions (Sathayea and Reddy, 2013). Among the tropical Asian countries, India has a high potential C content in forest vegetation. India along with Indonesia and Myanmar accounts for about 70% of total carbon pools in tropical Asian forests (Brown et al., 1993; FAO, 2010). Several studies have been published on the Indian forest carbon pool (Chhabra et al., 2002a,b; Kishwan et al., 2009; Haripriya, 2000) and also on emission from deforestation and land use changes (Chhabra and Dadhwal, 2004; Kaul et al., 2009). Growing stock from forest inventories is the primary input for C estimates. Based on the various approaches, the current estimates of Indian forest C pools are in the range of 2000–4400 tera gram carbon (TgC) (Chhabra et al., 2002a,b). The estimated Indian forest carbon densities for the recent periods (past 2 decades) range from 50 to 65 Mg ha−1 (Kishwan et al., 2009; Lal and Singh, 2000). Using C densities based on ecological studies and remote sensing (RS), aggregated country level forest C pool was estimated as 4179 TgC for 1986 (Ravindranath et al., 1997), 3321.7 TgC and 3117 TgC for 1980 and 1990, respectively (Dadhwal et al., 1998). Using field inventory of growing stock and standard biomass expansion and conversion factors, forest C pool was estimated as 3978 TgC and 4071 TgC for 1982 and 1991, respectively (Dadhwal and Shah, 1997), and 4341.8 TgC for 1993 (Chhabra et al., 2002b). Several studies have estimated spatial forest biomass using remote sensing approaches over different regions of India during past one decade (Jha et al., 2015; Thumaty et al., 2016; Devagiri et al., 2013). These studies used satellite derived parameters along with field inventory to develop spectral model for forest biomass estimation. Pargal et al. (2017) and Reddy et al. (2016) used texture metrics derived from high resolution remote sensing data to spatially map the forest above ground biomass. Véga et al. (2015) used airborne remote sensing data for above ground biomass estimation of complex tropical forest in India.

1.3. Spatial distribution of Indian forest India is the seventh-largest country in the world, with a total area of 328.7 million hectare (Mha).The country is situated north of the equator between 8°4′ and 37°6′ north latitude and 68°7′ and 97°25′ east longitude. India’s geographical area constitutes 2.4% of the world land area and about 1.7% of the global forests (FAO, 2010), while supporting 16% of the world’s human population. Using historical data, Richard and Flint (1994) estimated India's forest area in 1880 and 1980 as 102.68 and 64.6million ha, respectively. From 1981–83 onwards, the Forest Survey of India (FSI) has mapped and monitored India's forests on a biennial basis using satellite data (FSI, 1987). All lands, more than 1 ha in area, with a tree canopy density of more than 10 percent are considered as forest cover by FSI. 2. Materials and methods The present study aims to develop specially explicit estimates of forest C pool at 5 km grid level (for 1994 and 2010) using field inventory based growing stock (GS) volume and remote sensing based forest area. The growing stock was estimated using species specific and regional volumetric equations (FSI, 1996). The growing stock was converted to carbon using wood density (WD), biomass expansion factor (BEF) and carbon correction factor.

1.2. Forest C for Indian forests as part of global estimates Table 1 presents the mean C densities and forest carbon stock for India based on the studies of global biomass estimation. All values include below and aboveground forest carbon. Houghton (1999); DeFries et al. (2002) distinguish forest types within region, whereas Brown (1997))/ Achard et al. (2004) distinguish only by region. The estimation procedure of Houghton (1991) is mostly based on compilation of harvest measurement from ecological studies.

2.1. Remote-sensing based forest area Fourteen assessments of India’s forest have been published by the Forest Survey of India (FSI 1987, 1989, 1991, 1993, 1995, 1997, 1999, 2001, 2003, 2005, 2009, 2011, 2013, 2015). These reports provide data on forest area under three crown density classes, viz, very dense forest (D1) with more than 70% canopy density, moderately dense forest (D2) with canopy density between 40% and 70% and open forest (D3) with canopy density between 10% and 40% (FSI, 1995a). Canopy density is defined as the proportion of the forest floor covered by the vertical projection of the tree crowns (Jennings et al., 1999). For the purpose of this study, the area under very dense and moderately dense class were merged under dense category. Remote sensing based forest area under dense, and open forest were used for forest C estimation. Forest area under dense and open forest were obtained from forest canopy density maps developed by Forest Survey of India (FSI, 1995a, 2009) for the years 1994 and 2010 respectively. In 1995 assessment of forest cover, Indian Remote Sensing Satellite product IRS–1 B LISS II with spatial resolution of 36.25 m were used.

Table 1 Average C densities and Carbon stock for India based on global estimates in million gram carbon per hectare (Mg ha−1). Forest type/regions Tropical Asia

Houghton (1999); DeFries et al. (2002)

Brown (1997); Achard et al. (2004)

Gibbs and brown (2007)

IPCC (2006)

Tropical forest C stock estimate using biomass average approach(Mg ha−1) All forests – 151 – – Tropical equatorial 250 – 164 180/ forest 225* Tropical seasonal 150 – 142 105/169 forest Tropical dry forest – – 120 78/96 National level forest biomass carbon estimates for Indian forest based on global studies (TgC) Indian Forests 8997 7333 8560 5085 * values here are for continental ad insular Southeast Asia, respectively.

743



cell. C density was calculated on the total cell carbon. Importance value index (IVI) was computed for tree species at the state level. Importance value index is measure of how dominant a species is in a given forest stand. Species and region specific volume equations and wood density were used for the twenty most important species at the state level while pooled volume equations were employed for the others (Table 2). 173 species specific volumetric equations and 193 wood densities were used for plot biomass estimation. The allometric equations used in volume estimations are designed for the diameter range 10–100 cm. A regression model was established between Basal Area density (BA) and biomass density of trees with diameter more than 10 cm. Regression equations were used to calculate biomass stock of trees less than 10 cm DBH (Fig. 4). Plot aboveground biomass estimated from field inventory data for the year 2010 ranged from 1.02 Mg ha−1 to 621.75 Mg ha−1 with an average biomass density of 84.37 Mg ha−1. Methodology is shown in Fig. 3 and 4. Tree volume obtained by volumetric equation was converted to biomass using biomass expansion factor and wood density. The value of 0.47 has been used for the carbon fraction of dry matter

Visual interpretation and digital classification were used for forest cover assessment. Resourcesat-1 (also known as IRS-P6) LISS III data with spatial resolution of 23.5 m were used for forest cover assessment in 2009 using digital classification technique. 2.2. Field inventory data The growing stock information for 1994 was computed from a national field inventory carried out by FSI (FSI, 1995b). The approach divides the country into 0.042 × 0.0420. 10000 forested grids were selected across the country and in each of those cells, two plots of 0.1 ha (square plot, 31.62 × 31.62m) were inventoried. About 20,000 plots were inventoried across the country. The field inventory report provides total growing stock, volume density, basal area density, forest strata, stem count, stem density, incidences of fires, etc. (FSI, 1995a). Field inventory data collected in 2009–10 as part of a national forest inventory carried out for the ISRO-GBP National Carbon Project (Dadhwal et al., 2009) were used. The inventory used a design based approach to select sample sites of 250 × 250 m on the basis of spatial data (based on forest types, forest canopy density and NDVI obtained through satellite data). Clustered sampling approach was adopted for field sampling and therefore, at each site four sample of 0.1 ha (conventional plot size) were inventoried. Field Sampling design for 2010 is shown in Fig. 1. A total of 6028 plot of 0.1 ha size is used for the current analysis for the 2010 time period. GBH (girth at breast height, 1.3 m above ground) of all trees with GBH > 10 cm (DBH > 3.18 cm) were measured using measuring tape during field inventory. Field inventory data for 2840 sample plots in the forests of Western Ghats (a mountain range that runs parallel to the western cost of the Indian peninsula) was also integrated (Roy et al., 2012). Coordinates of all plots were recorded using GPS. Distribution of sample plots (sample plots used in 2010 estimates) over forest canopy density map and Agro ecological sub regions is shown in Fig. 2.

C = 0.47 × biomass (Aalde et al., 2006) Five km grid level C values for the year 2010 were computed as a product of C density and the total forest area of the 5 km cell for the respective year. Average C densities for open and dense forest were obtained for each zone by intersecting the field plots with agro-eco zone and canopy density map. A root correction factor of 1.16 (total carbon = 1.16 x above ground carbon) was used to include the below ground carbon in carbon density estimates (Chhabra et al., 2002a; Hall and Uhlig, 1991). Forest C density was estimated using the following equation: CD = GS × WD × CC × RC × BEF

(1) −1

Where CD is forest C density (Mg ha ), GS the growing stock (m3 ha−1), WD the density of wood (Mg m−3), RC the root correction factor = 1.16,CC the carbon content of the wood = 0.47 and BEF the biomass expansion factor (given in Table 2). District level C pools estimated in Chhabra et al. (2002a) were used to obtain 5 km level carbon pools for year 1994. Dense and open forest densities for the year 1994 were obtained by using ratio of 1.8 for dense to open forest C density. Dense and open forest areas were obtained from FSI, 1995a assessment of forest cover.

2.3. Estimation of forest carbon Plot biomass was estimated for 2010, using species and region specific volume equations and wood density and biomass expansion factor (Dadhwal et al., 2009). Forest C density was estimated at a 5 km grid cell level, by using a stratify and multiply approach (Goetz et al., 2009). The forest cover density map (FSI, 2009) was intersected with 5 km grid to obtain the forest area in two forest canopy density classes (Very dense and moderate dense classes in the FSI map were combined to create a dense canopy density strata) at each 5 km cell. Biomass density averages for dense and open forests (average of plot biomass in dense and open forest) were calculated for each of India’s 60 agro ecological sub regions (Velayutham et al., 1999) using field inventory plots categorized into dense or open forest strata on the basis of forest density map. Area under dense and open forest at each 5 km cell were multiplied with average biomass density for the dense and open forest classes respectively to estimate the total biomass in that particular 5 km

District Biomass Carbon Pool = PD × forest area, District Biomass (A1 + A2)) D1 = 1.8 × D2,

Carbon

Density = ((A1 × D1 + A2 × D2)/ (2) (3)

A1 and A2 are district level dense and open forest area, D1 and D2 are dense and open C densities. A1 and A2 were obtained from FSI 1994 reports. Eq. (2) and (3) were used to estimate D1 and D2 for each district. Fig. 1. Sampling design for 2010 field inventory data. a) 250 m site with four 0.1 ha plots, b) 0.1 ha plot for enumerating trees in forest.

744



Fig. 2. Field sample plots overlaid over forest canopy density map and Agro ecological sub regions of India.

3. Results

District level open and dense forest carbon densities for 1994 were used to estimate forest C pools in 5 km cell. Methodology for C estimation at 5 km cell is given in Fig. 3. 5 Km grid Biomass Carbon = a1 × D1 + a2 × D2Where, a1 and a2 are dense and open forest cover within 5 km cell.a1 and a2were obtained from FSI 1995 report.

3.1. Remote-sensing based forest area There was a net increase of 5.74 Mha in national forest area i.e., from 63.34 Mha in 1994–69.08 Mha in 2009. Increase in dense and open category forest was 2.09 Mha and 3.65 Mha, respectively. State wise comparison of dense and open forest area for 1994 and 2009 is provided in Table 3. State wise analysis of forest cover showed that decrease in dense forest for the states Assam, Gujarat, Himachal Pradesh, Manipur, Nagaland and Odisha is followed by increase in open forest, while decrease in open forest is followed by increase in dense forest for Meghalaya state. States of Haryana, Himachal Pradesh and Nagaland showed an overall decrease in forest cover over the 1994–2009 time period. Net increase of more than 0.4 Mha is observed in 5 states (Arunachal Pradesh, Chhattisgarh, Karnataka, Madhya Pradesh and Maharashtra). Forest area increase by more than 0.2 Mha was observed in 13 states and decrease in forest area was observed in 3 states.

2.4. Uncertainty analysis An uncertainty analysis was performed to evaluate the error of estimated AGB in a 5 km grid cell. For each ground plot, the ground measured AGB and the estimated AGB using stratify and multiply approach were used to estimate the estimation error. RMSE was calculated for all ground plots. The uncertainty in RMSE was reported as the 95% confidence interval (CI). To estimate CI, bootstrapping sampling approach was used to estimate RMSE for 1000 simulations (MonteCarlo approximation). These 1000 RMSE estimates formed a probability distribution function representing mean and variance of the RMSE of estimated AGB. Finally, 95% CI was characterized using t-distribution. Error was computed as agro ecological sub regions and also at national level.

3.2. Forest carbon density estimates Agro-ecological sub region strata average for open forest ranged 745



Table 2 Pooled Volume equations, wood density(values given in Mg m−3) and biomass expansion factors used in current study are as follows. State

WD(Mean)

BEF

Pooled Volume Equations

Andhra Pradesh Arunachal Pradesh Assam Bihar Chhattisgarh Delhi Goa Gujarat Haryana Himachal Pradesh Jammu & Kashmir Jharkhand Karnataka Kerala Maharashtra Manipur Meghalaya Mizoram Madhya Pradesh Nagaland Odisha Punjab Rajasthan Sikkim Tamil Nadu Tripura Telangana Uttar Pradesh Uttarakhand West Bengal

0.63 0.63 0.542 0.63 0.633 0.63 0.603 0.629 0.63 0.602 0.602 0.633 0.603 0.63 0.611 0.51 0.559 0.52 0.619 0.51 0.633 0.63 0.629 0.75 0.8 0.51 0.63 0.619 0.619 0.63

1.59 1.58 1.58 1.58 1.59 1.59 1.59 1.59 1.59 1.52 1.51 1.58 1.59 1.59 1.58 1.57 1.53 1.57 1.59 1.58 1.59 1.57 1.59 1.58 1.59 1.59 1.59 1.56 1.56 1.58

V = 0.088183 − 1.490948D + 8.984266D2 V = 0.15958 − 1.57976D + 8.25014D2 − 0.48518D3 V = 0.11079 − 1.81103D + 11.4132D2 + 0.38528D3 V = 0.025584 − 0.89224D + 9.5879D2 V = 0.0697 − 1.4597D + 11.79933D2 − 2.35397D3 √V = − 0.00342 − 0.0922D + 2.28178D2 + 9.46641D3 V = 0.058 + 4.598D2 V = − 0.15397 + 2.724109D √V = − 0.00342 − 0.0922D + 2.28178D2 + 9.46641D3 V = 0.193297 − 2.267002D + 10.679492D2 V = 0.23781 − 2.09431Dv+ 7.78268D2 V = 0.088074 − 1.449236D + 8.760534D2 V = 0.058 + 4.598D2 V = 0.058 + 4.598D2 V = 0.040841 − 0.88376D + 7.25224D2 + 1.34817D3 V = 0.68 + 5.049937D − 2.77092√D V = 0.70958 + 13.03244D − 72.25358D2 + 147.4303D3 V = 0.68 + 5.049937D − 2.77092√D V = 0.0697 − 1.4597D + 11.79933D2 − 2.35397D3 √V = − 0.2264 + 2.93587D V = 0.088074 − 1.449236D +v8.760534D2 √V = − 0.00342 − 0.0922D + 2.28178D2 + 9.46641D3 V = 0.081467 − 1.06366D + 6.452918D2 V = 0.3555 − 3.700D + 12.590D2 V = 0.058 + 4.598D2 √V = − 0.2264 + 2.93587D V = 0.088183 − 1.490948D + 8.984266D2 V = 0.17553 + 7.94663D2 − 7.1434√D V = 0.17553 + 7.94663D2 − 7.1434√D V = 0.15958 − 1.57976D + 8.25014D2 − 0.48518D3

*Biomass expansion factor, wood density and volume equations were derived from Kaul et al. (2011), FRI, 1996 and FSI, 1996 respectively. (Where, D = DBH and V=Volume).

from 1.49 to 172.63 Mg ha−1 with a mean of 53.98 Mg ha−1, while for dense forest it ranged from 5.97 to 364.52 Mg ha−1 with a mean biomass density of 90.25 Mg ha−1. State level C pools are shown in the Table 4. Total C for every state is given as sum of C of individual 5 km cells within the corresponding state. National forest carbon estimated for 1994 and 2010 are 3911.78 TgC (2895.28 TgC in dense forest and 1016.50 TgC in open forest) and 4368.03 TgC (3176.48 TgC in dense forest and 1191.55 TgC in open forest). These estimates showed a net increase of 456.25 TgC in 16 years with an annual increment of 25.52 TgC year−1. Mean forest carbon density increased from 61.14 Mg ha−1 in 1994 to 64.08 Mg ha−1 in 2010. C densities for dense and open forest in 1994 estimated as 77.08 and 38.47 Mg ha−1 with total C pools of 2895.28 TgC and 1016.50 TgC, which increased to 80.24 Mg ha−1 and 41.69 Mg ha−1 with total C pools of 3176.48 TgC and 1191.55 TgC, from 1994 to 2010. State level change analysis showed maximum increase of 176.80 TgC in Uttarakhand state and maximum decrease of 110.40 TgC in Jammu and Kashmir. Maximum forest C for 2010 was estimated in Arunachal Pradesh, followed by Uttarakhand, Madhya Pradesh, Chhattisgarh and Jammu and Kashmir as 695.68, 373.14, 321.67, 296.72 and 293.86 TgC, respectively, while in 1994, maximum forest C was estimated in Arunachal Pradesh, Jammu and Kashmir, Madhya Pradesh, Chhattisgarh, and Assam states as 613.03, 404.26, 338.47, 297.96, and 243.39 TgC respectively. Spatial distribution of forest C density for 1994 and 2010 is shown in Fig. 5.

relatively low sampling intensity. At national level, error of forest C estimation is computed as 33.57% (95% confidence interval of 32.77–34.17%). Forest C in Indian forest is estimated as 4368.03 ± 33.57% in 2010. Relative error estimated at Agro ecological sub regions were converted to 5 km grid level C density estimation error by multiplying with grid C density values. Error in forest C estimation is shown at 5 km spatial resolution in Fig. 6. Forest C density error surface indicate low estimation error in central Indian deciduous forests, which is attributed due to higher sampling intensity and medium range carbon density. Moderate estimation errors were observed in regions with medium C density range and lower sampling intensity. Higher estimation error was observed in Western Himalayan, Eastern Himalayan and, Western Ghats region due to higher C density ranges and high local variation over these regions. Uncertainty analysis suggests that the to achieve less uncertainty in the estimation in medium range C density regions required higher sampling intensity. Uncertainty for 1994 estimates could not be evaluated due to the lack of access to detailed tree level data. Best available district level forest C stock were downscaled to 5 km grid level using forest area statistics in two major categories open and dense forest. 4. Discussion 4.1. Forest area and forest carbon density estimates India's forest cover increased by 5.74 Mha during the 16 years of study period. Despite increasing populations and large dependence of rural communities on forest resources for their various needs, India has successfully managed to increase its forest cover over past 2 decades through various conservation and afforestation efforts by government. India's forest cover grew at 0.22% per year over 1990–2000, and at the rate of 0.46% per year over 2000–2010 (FAO, 2010). Along with the increase in forest cover from 1994 to 2010, there has

3.3. Uncertainty in forest carbon estimates An uncertainty analysis was performed to evaluate the error and the uncertainty of the error of estimated carbon at agro ecological sub regions and also at national level. Relative RMSE estimates ranged between 7 to 44% at agro ecological sub region level. Higher prediction error in some of the agro ecological sub region is attributed due to 746



Fig. 3. Methodology for plot biomass estimations using field inventory data and volumetric equations, and 5 km grid biomass estimation using stratify and multiply approach.

Fig. 4. Biomass estimation for tree diameter 3.18–10 cm for year 2010.a) regression between basal area and biomass for DBH > 10 cm and b) regression model for biomass and basal area for DBH 3.18–10 cm.

overall forest health. Growing stock, forest biomass and forest C and its spatial distribution needs to be considered as equally important. We present the spatial distribution of forest carbon density and uncertainty in the estimation at 5 km spatial resolution. Central Indian deciduous forests showed a relatively stable pattern and forest C density has not shown significant change during study period (Fig. 5).

been increase in the mean biomass and carbon density per unit area i.e., from 61.14 to 64.08. It has been demonstrated that the improvement in forest C density is possible through the conversion of open forest to dense and very dense forest. Increase in forest cover and forest C density both contributed to increase in C pools, which indicated that forest cover alone cannot be considered as indicator of improvement in 747



Table 3 State wise comparison of forest area in open and dense forest. State

Andhra Pradesh Arunachal Pradesh Assam Bihar Chhattisgarh Delhi Goa Gujarat Haryana Himachal Pradesh Jammu & Kashmir Jharkhand Karnataka Kerala Madhya Pradesh Maharashtra Manipur Meghalaya Mizoram Nagaland Odisha Punjab Rajasthan Sikkim Tamil Nadu Telangana Tripura Uttar Pradesh Uttarakhand West Bengal

Area −2009 (km2)

Area-1994 (km2)

Change:1994–2009 (km2)

Dense

Open

Total

Dense

Open

Total

Dense

Open

Total

10093.6 50295.86 15085.80 2625.33 35087.50 49.99 794.80 5876.61 378.49 10289.02 13098.63 10271.74 21653.15 8193.99 40823.55 26406.03 6442.65 6561.91 5919.46 10775.56 31293.78 502.55 4826.32 2508.54 11721.00 12422.74 4500.90 4902.85 17564.02 5157.76

10406.42 12177.20 12708.58 2202.63 16104.60 67.77 799.60 8289.61 1072.52 5141.00 8607.89 10294.26 10781.16 4383.18 33092.74 17532.63 9596.58 8895.72 11852.19 3785.25 15001.10 564.07 11345.14 684.43 9199.25 7593.86 2055.44 6763.96 5485.35 4838.66

20500.02 62473.05 27794.37 4827.96 51192.10 117.76 1594.40 14166.22 1451.01 15430.01 21706.52 20566.00 32434.31 12577.17 73916.28 43938.66 16039.23 15457.63 17771.64 14560.81 46294.88 1066.62 16171.46 3192.97 20920.25 20016.6 6556.34 11666.81 23049.37 9996.42

10743.31 52930.78 13798.27 3766.59 39787.01 47.48 790.08 5609.08 344.06 10024.15 13349.75 12818.43 23107.04 9359.659 42720.47 28044.07 5939.05 9037.08 6731.84 5944.941 27517.86 581.64 4883.95 2861.9 13604.58 13611.95 4711.22 6577.63 19111.01 7504.63

9229.29 15254.39 16131.83 3449.29 16818.49 71.70 845.30 8563.81 1058.85 5309.99 9271.36 10350.23 14721.81 5863.39 35808.28 20602.08 10799.52 6715.76 12483.59 6946.17 19897.38 773.18 12147.08 761.38 9908.12 9273.28 3060.61 8613.94 5660.12 5365.96

19982.59 68185.18 29930.10 7215.88 56605.50 119.19 1635.39 14172.89 1402.92 15334.15 22621.11 23168.66 37828.85 15223.04 78528.75 48646.15 16738.57 15752.85 19215.43 12891.12 47415.24 1354.81 17031.03 3623.32 23512.71 22885.24 7771.83 15191.57 24771.13 12870.59

659.71 2634.93 −1287.52 1141.26 4699.52 −2.50 −4.72 −267.52 −34.43 −264.86 251.12 2546.68 1453.89 1165.66 1896.92 1638.04 −503.60 2475.18 812.39 −4830.62 −3775.92 79.09 57.63 353.41 1883.59 1189.21 210.32 1674.78 1546.99 2346.87

−1177.13 3077.20 3423.25 1246.67 713.89 3.93 45.70 274.20 −13.67 169.00 663.47 55.97 3940.65 1480.22 2715.54 3069.45 1202.94 −2179.96 631.41 3160.92 4896.28 209.10 801.94 76.95 708.88 1679.42 1005.17 1849.98 174.77 527.30

−517.42 5712.13 2135.73 2387.93 5413.41 1.43 40.98 6.68 −48.09 −95.86 914.58 2602.66 5394.54 2645.88 4612.46 4707.49 699.34 295.22 1443.79 −1669.70 1120.36 288.18 859.57 430.35 2592.46 2868.64 1215.49 3524.76 1721.76 2874.18

*Area under dense and open forest were derived from FSI, 1995a and FSI, 2009.

Table 4 Forest Carbon Stock Change statistics 1994−2010. State

Andhra Pradesh Arunachal Pradesh Assam Bihar Chhattisgarh Delhi Goa Gujarat Haryana Himachal Pradesh Jammu & Kashmir Jharkhand Karnataka Kerala Madhya Pradesh Maharashtra Manipur Meghalaya Mizoram Nagaland Odisha Punjab Rajasthan Sikkim Tamil Nadu Telangana Tripura Uttar Pradesh Uttarakhand West Bengal Total

Carbon 1994 (TgC)

Carbon 2010 (TgC)

Change (1994−2010)

Dense

Open

C-1994

Dense

Open

C-2012

Dense

Open

Total

80.24 537.95 165.16 9.61 235.11 0.15 8.43 31.55 1.16 150.73 291.68 40.16 176.97 66.52 230.19 152.64 39.23 48.87 25.12 69.20 151.09 0.41 10.04 23.58 48.63 64.21 12.21 36.88 166.09 21.48 2895.28

31.53 75.08 78.23 4.36 62.85 0.12 4.86 30.34 2.00 42.55 112.58 24.37 43.07 21.61 108.28 59.47 34.97 39.49 28.63 14.31 44.68 0.30 12.23 3.80 22.62 41.75 3.30 29.77 30.26 9.10 1016.50

111.78 613.03 243.39 13.97 297.96 0.27 13.29 61.89 3.15 193.28 404.26 64.53 220.04 88.14 338.47 212.12 74.19 88.36 53.75 83.51 195.77 0.71 22.27 27.38 71.25 105.95 15.51 66.65 196.35 30.58 3911.78

57.02 596.80 140.72 39.32 230.49 0.24 6.56 17.33 3.01 120.54 178.90 66.72 189.57 78.00 221.76 185.30 29.24 84.85 33.08 53.10 182.99 8.30 13.24 29.32 82.37 60.53 24.67 53.34 332.87 56.26 3176.48

21.44 98.89 112.25 14.54 66.23 0.23 3.92 9.18 3.56 66.64 114.97 40.51 62.47 27.56 99.91 77.13 16.77 40.21 19.75 32.23 80.08 5.33 21.26 4.16 38.46 16.41 5.55 30.38 40.28 21.25 1191.55

78.53 695.68 252.97 53.86 296.72 0.47 10.47 26.51 6.57 187.17 293.86 107.23 252.04 105.57 321.67 262.44 46.01 125.05 52.83 85.33 263.07 13.63 34.50 33.49 120.83 76.94 30.22 83.72 373.15 77.51 4368.03

−23.22 58.84 −24.44 29.72 −4.63 0.09 −1.87 −14.22 1.85 −30.19 −112.78 26.57 12.60 11.48 −8.43 32.66 −9.99 35.98 7.97 −16.10 31.90 7.89 3.20 5.75 33.74 −3.68 12.46 16.46 166.78 34.78 281.20

−10.09 23.81 34.01 10.18 3.39 0.11 −0.94 −21.16 1.57 24.08 2.39 16.14 19.40 5.95 −8.37 17.66 −18.19 0.72 −8.88 17.92 35.39 5.03 9.02 0.36 15.84 −25.34 2.25 0.61 10.02 12.16 175.05

−33.27 82.65 9.58 39.90 −1.24 0.20 −2.81 −35.37 3.41 −6.11 −110.40 42.70 32.00 17.43 −16.80 50.32 −28.18 36.70 −0.91 1.83 67.30 12.92 12.22 6.11 49.58 −29.01 14.70 17.07 176.80 46.94 456.25

748



Fig. 5. Forest Carbon density surface a) 1994 and b) 2010.[Note: Analysis for Andaman & Nicobar Island and Lakshadweep is not included].

Although, forest C density remains in the range of 55–110 Mg ha−1 in Western Ghats (Indian biodiversity hot-spot region (Myers et al., 2000)) however, significant decrease in forest C density is observed in patches in this region. The western Himalayan region includes the states of Jammu and Kashmir, Himachal Pradesh, Uttarakhand and parts of Punjab. C density for Himachal Pradesh and Uttarakhand was in the range of 55–150 Mg ha−1, however, some parts of Jammu and Kashmir showed C density higher than 150 Mg ha−1 in 1994. Significant decrease in forest carbon was observed in the state of Jammu and Kashmir, total C pools for the state decrease from 404.26 TgC 293.86 TgC (decreased by 110.40 TgC in 16 years). Increase in forest C density was observed in Uttarakhand state wherever the areas had C density above 110 Mg ha−1 in 2010. Total forest C pools almost doubled over the study period for the state (196.35 373.17 TgC).This increase in C density may be attributed to the prevalence of dense forests and less disturbance as these forests are situated on less populated mid and high hills of western Himalaya. Eastern Himalayan region consists of Sikkim, parts of West Bengal and Arunachal Pradesh. This region is characterized by hills, mountains and near tropical to alpine climatic conditions. The forest C density remains in the range of 55–150 Mg ha−1 for this region over the study period, increase in total forest C pool in region is mainly due to the increase in forest cover. Arunachal Pradesh contains the largest forest C pools of the states. Forest C density in the state remains almost same over the study period; increase of 82.65 TgC is on account of the increase in forest area by 5712 km2. The earlier estimates of Indian forest C pools are in the range of 2000–4400 TgC, however majority of them lie between 3100 and 4400 TgC (Chhabra et al., 2002a). We have estimated the Indian C Pools as 3911.78 TgC and 4368.03 TgC in 1994 and 2010 respectively, which lie in the range of earlier studies. The difference, however little, in estimates maybe because, 1) earlier estimates did not include biomass expansion factors, 2) factor used for biomass to carbon conversion and 3) increment in biomass due to increased age structure of forest and increase in total forest cover. Most of the earlier estimates used single biomass average at strata level for both dense and open forest, which

results in underestimation of forest C, since dense forest occupies large part of Indian forests.

4.2. Comparison with other estimates National level forest biomass carbon estimates for Indian forest based on global studies are in the range 5085–8897 TgC (Table 1). Forest carbon stock for the country was calculated by using average forest biomass carbon values in combination with a satellite based global land cover map for the year 2000 stratified by the FAO forest ecological zone map (FAO, 2000). Nearly all estimates of global forest biomass Care based on biomass average datasets where a single representative value of forest carbon per unit area (e.g. tonnes of C per hectare) was applied to broad forest categories. Forest biomass C estimates for Indian forests based on global studies are much higher than present study, which may be attributed to higher biomass averages used in global studies. The available biomass datasets are based on compilation of individual published studies and also there is a tendency for researchers to select dense forests in a locale specific work or to avoid disturbed forests which result in overestimation of forest biomass C pools. IPCC 2006 default values based Tier 1 estimates of national C pools for Indian forest is estimated as 5085 TgC. Information on biomass is available from many sources including insitu measurements, national forest inventories, model outputs and regional satellite products. Global Forest Resource Assessment (FRA) produced by Food and Agriculture Organization of the United Nations (FAO, 2005) is one of the few available maps. Kindermann et al. (2008) downscaled the known country level biomass (aggregated result of the FRA 2005) to a half degree global spatial datasets based on two factors NPP and human impacts. Fig. 7a shows the comparison of current estimates with FAO statistics based spatial estimates by Kindermann et al. (2008). Kindermann's estimates are based on 2005 FAO statistics and show good correlation (R2 = 0.715) with 2010 estimates of current study but AGB density values are relatively low. Since, Kindermann downscaled the country level biomass to the half degree spatial resolution, one possible reason for underestimation of AGB density is the low spatial resolution (full pixel may not have been covered by forest). 749



Fig. 6. Forest carbon density error surface at 5 km grid.

waveform characteristics to AGB from 283 calibration field plots located under GLAS footprints. Since Baccini et al. (2012) used 2007–08 field datasets, AGB density estimates are in the range of 2010 estimates of current study. Baccini et al. (2012) provides spatial estimates of AGB density for tropical forests of India and shows a significant correlation (R2 = 0.635) with current estimates because field observations of similar forest types is used in study. Fig. 7 shows the comparison of current estimates with earlier estimates for Indian forests based on global studies. Comparison of current estimates with available global estimates shows that current estimates are in the range of earlier estimates, but show spatial inconsistency which may be attributed to methodology adopted, less number of field observations used and generalizations of field plots over large areas in global estimates. Saatchi et al. (2011) and Baccini et al. (2012) used less number of field observations covered over smaller regions to extrapolate AGB density over larger area. Current estimates used intense field sampling datasets and were thus able to pick the high variations in forest C density over Indian forests. Inclusion of GLAS and other remote sensing data with such a high field inventory datasets may improve estimates further and provides AGB estimates at higher resolutions.

Ruesch and Gibbs (2008) followed IPCC Tier-1 methods for estimating vegetation carbon stock using globally consistent default values provided for aboveground biomass (IPCC, 2006). Belowground biomass (root) carbon stocks were added using the IPCC root to shoot ratios for each vegetation type. Ruesch and Gibbs presents the biomass carbon density for 2000 and the shows a significant correlation (R2 = 0.736) with 1994 estimates of current study. Since these estimates are based on IPCC default values which are generalized over larger areas and might have resulted into the apparent overestimation in forest biomass carbon in higher biomass regions, and also not able to pick the variation in high biomass regions. Saatchi et al. (2011) generated benchmark map of forest carbon stock in tropical regions across three continents using GLAS data. Saatchi et al. (2011) derived three separate continental equations relating Lorey’s height to AGB using a set of 493 field plots. Saatchi's AGB density estimates are in the range of current estimate for year 2010. Saatchi's estimates for Indian forest are based on 120 field sampling plots over Southeast Asia, which do not offer adequate coverage of all forest type and density classes. Saatchi's estimates perform well in high biomass regions but unable to pick the low biomass regions (which are thus over estimated) and are spatially inconsistent with current estimates in low biomass regions. Baccini et al. (2012) builds a model directly relating GLAS 750



Fig. 7. Comparison of C density estimates with (a) Kindermann's estimates for 2005, (b) Ruesch and Gibbs estimates for 2000, (c) Saatchi's estimates for 2005 and (d) Baccini's estimates for 2007–08 (biomass from different maps were plotted at randomly selected points).

5. Conclusion

VCP component study of the NCP and also the Department of Biotechnology, Government of India and University of Agricultural Sciences, Bangluru for sharing of the field inventory data for the Western Ghats.

In this present study, an attempt was made to estimate forest C stock and its spatial distribution in Indian forests, because estimation of forest C is a critical step in quantifying Carbon stocks and fluxes from forests. Forest growing stock inventories are valuable source of data for estimating forest above ground biomass and a basis for estimation of Carbon source and sink. Our results show that Indian forests are sequestering Carbon. Mean forest carbon density has increased from 61.14 Mg ha−1 in 1994 to 64.08 Mg ha−1 in 2010. National forest C estimated for 1994 and 2010 as 3911.78 and 4368.03 TgC, shows a net increase of 456.26 TgC in 16 years. Growing populations increases the biomass demand and pressure on Indian forest but, the conversion of forest lands for other uses has actually declined since 1980. This is mainly due to the effective implementation of measures to control deforestation of natural forests. The study showed that forest C density has changed over the study period in Indian forests which indicate that forest cover alone cannot be considered as single indicator factor for forest development, forest biomass and Carbon storage need to be considered as equally important parameters. The present study highlighted the regional differences in forest C density in Indian forest. Forest C densities along with RS data can be used as inputs to the carbon models to understand the role of forests in atmospheric C fluxes. The understanding of spatial distribution of forest C is important for sustainable development and conservation of forests and for understanding the contribution of forest to carbon emissions and their potential for carbon sequestration. The use of remote sensing and GIS techniques coupled with ground inventory information of forests may improve the methodology for forest C estimation.

References Aalde, H., Gonzalez, P., Gytarsky, M., Krug, T., Kurz, W.A., Ogle, S., Raison, J., Schoene, D., Ravindranath, N.H., Elhassan, N.G., Heath, L.S., Higuchi, N., Kainja, S., Matsumoto, M., Sánchez, M.J.S., Somogyi, Z., 2006. Forest land. In Intergovernmental Panel on Climate Change. National Greenhouse Gas Inventory Guidelines. Institute for Global Environmental Strategies, Hayama, Japan. Achard, F., Eva, H.D., Mayaux, P., Stibig, H.-J., Belward, A., 2004. Improved estimates of net carbon emissions from land cover change in the tropics for the 1990. Glob. Biogeochem. Cycles 18. Baccini, A., Goetz, S.J., Walker, W.S., Laporte, N.T., Sun, M., Sulla-Menashe, D., Hackler, J., Beck, P.S.a, Dubayah, R., Friedl, M.a., Samanta, S., Houghton, R.A., 2012. Estimated carbon dioxide emissions from tropical deforestation improved by carbondensity maps. Nat. Clim. Change 2, 182–185. Brown, S., Gaston, G., 1995. Use of forest inventories and geographic information systems to estimate biomass density of tropical forests: application to tropical Africa. Environ. Monit. Assess. 38, 157–168. Brown, S., Iverson, L.R., Prasad, A., Liu, D., 1993. Geographical distributions of carbon in biomass and soils of tropical Asian forests. Geocarto Int. 8, 45–59. Brown, S., Sathaye, J., Cannell, M., Kauppi, P.E., 1996. Mitigation of carbon emissions to the atmosphere by forest management. Commonw. For. Rev. 80–91. Brown, S., 1997. Estimating biomass and biomass change of tropical forests: a primer. Food Agric. Org. 134. Chhabra, A., Dadhwal, V.K., 2004. Assessment of major pools and fluxes of carbon in Indian forests. Clim. Change. 64, 341–360. Chhabra, A., Palria, S., Dadhwal, V.K., 2002a. Spatial distribution of phytomass carbon in Indian forests. Glob. Change Biol. 8, 1230–1239. Chhabra, A., Palria, S., Dadhwal, V.K., 2002b. Growing stock-based forest biomass estimate for India. Biomass Bioenergy 22, 187–194. Ciais, P., Sabine, C., Bala, G., Bopp, L., Brovkin, V., Canadell, J., Chhabra, A., DeFries, R., Galloway, J., Heimann, M., Jones, C., 2013. Carbon and other biogeochemical cycles. 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, pp. 465–570. Dadhwal, V.K., Shah, A., 1997. Recent changes in forest phytomass carbon pool in India estimated using growing stock and remote sensing based forest inventories. J. Trop. For. 13, 188–192. Dadhwal, V.K., Pandya, N., Vora, A.B., 1998. Carbon cycle for Indian forest ecosystem: a preliminary estimate. Global Change Studies, Scientific Results from ISRO Geospherebiosph. Program. ISRO, India, pp. 411–429.

Acknowledgement This study is a part of the National Carbon Project (NCP) funded by Indian Space Research Organization, Government of India under its ISRO-Geosphere Biosphere Program. We acknowledge the PIs of the 751



Kaul, M., Dadhwal, V.K., Mohren, G.M.J., 2009. Land use change and net C flux in Indian forests. For. Ecol. Manage. 258, 100–108. Kaul, M., Mohren, G.M.J., Dadhwal, V.K., 2011. Phytomass carbon pool of trees and forests in India. Clim. Change 108, 243–259. Kindermann, G.E., Mccallum, I., Fritz, S., Obersteiner, M., 2008. A global forest growing stock, biomass and carbon map based on FAO statistics. Silva Fenn. 42, 387–396. Kishwan, J., Pandey, R., Dadhwal, V.K., 2009. India's Forest and Tree Cover: Contribution as a Carbon Sink. Indian Counc. For. Res. Educ. Lal, M., Singh, R., 2000. Carbon sequestration potential of Indian forests. Environ. Monit. Assess. 60, 315–327. Myers, N., Mittermeier, R.A., Mittermeier, C.G., Da Fonseca, G.A.B., Kent, J., 2000. Biodiversity hotspot for conservation priorities. Nature 403, 853–858. Noble, I., Bolin, B., Ravindranath, N.H., Verardo, D.J., Dokken, D.J., 2000. Land Use, Land-Use Change and Forestry. University Press, Cambridge. Olson, J.S., Watts, J.A., Allison, L.J., 1983. Carbon in Live Vegetation of Major World Ecosystems. Oak Ridge National Lab, TN, (USA) (No. DOE/NBB-0037). Pan, Y., Birdsey, R. a., Phillips, O.L., Jackson, R.B., 2013. The structure, distribution, and biomass of the world’s forests. Annu. Rev. Ecol. Evol. Syst. 44, 593–622. Pargal, S., Fararoda, R., Rajashekar, G., Balachandran, N., Réjou-Méchain, M., Barbier, N., Jha, C.S., Pélissier, R., Dadhwal, V.K., Couteron, P., 2017. Inverting aboveground Biomass–Canopy texture relationships in a landscape of forest mosaic in the western ghats of India using very high resolution cartosat imagery. Remote Sens. 9 (3), 228. Penman, J., Gytarsky, M., Hiraishi, T., Krug, T., Kruger, D., Pipatti, R., Buendia, L., Miwa, K., Ngara, T., Tanabe, K., Wagner, F., 2003. Good Practice Guidance for Land Use, Land-use Change and Forestry. Institute for Global Environmental Strategies. Ravindranath, N.H., Somashekhar, B.S., Gadgil, M., 1997. Carbon flow in Indian forests. Clim. Change 35, 297–320. Reddy, R.S., Rajashekar, G., Jha, C.S., Dadhwal, V.K., Pelissier, R., Couteron, P., 2016. Estimation of above ground biomass using texture metrics derived from IRS cartosat1 panchromatic data in evergreen forests of western ghats, India. J. Indian Soc. Remote Sens. 1–9. Richard, J.F., Flint, E.P., 1994. Historic Land Use and Carbon Estimates for South and Southeast Asia 1880–1980. Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory. Roy, P.S., Kushwaha, S.P.S., Murthy, M.S.R., Roy, A., Kushwaha, D., Reddy, C.S., Behera, M.D., Mathur, V.B., Padalia, H., Saran, S., 2012. Biodiversity Characterisation at Landscape Level: National Assessment. Indian institute of remote sensing, Dehradun, India. Ruesch, A., Gibbs, H.K., 2008. New IPCC Tier-1 global biomass carbon map for the year 2000. Saatchi, S.S., Harris, N.L., Brown, S., Lefsky, M., Mitchard, E.T.a., Salas, W., Zutta, B.R., Buermann, W., Lewis, S.L., Hagen, S., Petrova, S., White, L., Silman, M., Morel, A., 2011. Benchmark map of forest carbon stocks in tropical regions across three continents. Proc. Natl. Acad. Sci. U. S. A. 108, 9899–9904. Sathayea, J., Reddy, A., 2013. Integrating ecology and economy in India. Global Greenhouse Regime, Who Pays? Science, Economics and North-south Politics in the Climate Change Conventions. UNU, Tokio (Japan), pp. 191 (No. GTZ-844). Stocker, T.F., Qin, D., Plattner, G.-K., Tignor, M., Allen, S.K., Boschung, J., Nauels, A., Xia, Y., Bex, V., Midgley, P.M., 2013. Climate change 2013: the physical science basis. Intergovernmental Panel on Climate Change Working Group I Contribution to IPCC Fifth Assessment Report (AR5). Cambridge University Press, New York, NY, U. S. A. Tan, K., Piao, S., Peng, C., Fang, J., 2007. Satellite-based estimation of biomass carbon stocks for northeast China’s forests between 1982 and 1999. For. Ecol. Manage. 240, 114–121. Thumaty, K.C., Fararoda, R., Middinti, S., Gopalakrishnan, R., Jha, C.S., Dadhwal, V.K., 2016. Estimation of above ground biomass for central Indian deciduous forests using ALOS PALSAR L-band data. J. Indian Soc. Remote Sens. 44 (1), 31–39. Véga, C., Vepakomma, U., Morel, J., Bader, J.L., Rajashekar, G., Jha, C.S., Ferêt, J., Proisy, C., Pélissier, R., Dadhwal, V.K., 2015. Aboveground-biomass estimation of a complex tropical forest in India using lidar. Remote Sens. 7 (8), 10607–10625. Velayutham, M., Mandal, D.K., Mandal, C., Sehgal, J.L., 1999. Agro-ecological subregions of India for development and planning. NBSS Publ. 35, 452. Whittaker, R.H., Likens, G.E., 1973. Primary production: the biosphere and man. Hum. Ecol. 1, 357–369.

Dadhwal, V.K., Singh, S., Patil, P., 2009. Assessment of phytomass carbon pools in forest ecosystems in India. NNRMS Bull. DeFries, R.S., Houghton, R.A., Hansen, M.C., Field, C.B., Skole, D., Townshend, J., 2002. Carbon emissions from tropical deforestation and regrowth based on satellite observations for the 1980 and 1990. Proc. Natl. Acad. Sci. 99, 14256–14261. Devagiri, G.M., Money, S., Singh, S., Dadhawal, V.K., Patil, P., Khaple, A., Devakumar, A.S., Hubballi, S., 2013. Assessment of above ground biomass and carbon pool in different vegetation types of south western part of Karnataka, India using spectral modeling. Trop. Ecol. 54 (2), 149–165. Food and Agriculture Organization, 2000. Global Forest Resources Assessment 2000 Main Report. FAO, pp. 478 (For. Pap. 20). Food and Agriculture Organization, 2005. Global forest resources assessment 2005. Progress Towards Sustainable Forest Management. Food and Agriculture Organization, pp. 350 (FAO For. Pap. 147). Food and Agriculture Organization, 2010. Global forest resources assessment 2010. Main Report. Food and Agriculture Organization, pp. 378 (FAO For. Pap.). FRI, 1996. Indian Woods: Their Identification, Properties and Uses. Volume I-VI. Forest Research Institute, Ministry of Environment and Forests, Dehra Dun, India. FSI, 1987. The State of Forest Report 1987. Forest Survey of India, Govt. of India, Ministry of Environment & Forest, Dehradun (India). FSI, 1995a. The State of Forest Report 1995. Forest Survey of India, Govt. of India, Ministry of Environment & Forest, Dehradun (India). FSI, 1995b. Extent, composition, density, growing stock and annual increment of India's forests. Forest Survey of India. Ministry of Environment And Forests, Govt. of India, Dehradun. FSI, 1996. Volume Equations for Forests of India, Nepal and Bhutan Forest Survey of India. Ministry of Environment and Forests, Govt. of India, Dehradun. FSI, 2009. The State of Forest Report 2009 Forest Survey of India. Govt. of India, Ministry of Environment & Forest, Dehradun (India). Gibbs, H.K., Brown, S., 2007. Geographical Distribution of Biomass Carbon in Tropical Southeast Asian Forests an Updated Database for 2000. 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, 45023. Goetz, S.J., Baccini, A., Laporte, N.T., Johns, T., Walker, W., Kellndorfer, J., Houghton, R.A., Sun, M., 2009. Mapping and monitoring carbon stocks with satellite observations:a comparison of methods. Carbon Balance Manag. 4, 2. Hall, C.A.S., Uhlig, J., 1991. Refining estimates of carbon released from tropical land use change. Can. J. For. Res. 21 (1), 118–131. Haripriya, G.S., 2000. Estimates of biomass in Indian forests. Biomass Bioenergy 19, 245–258. Houghton, R.A., Hackler, J.L., Daniels, R.C., 1995. Continental Scale Estimates of the Biotic Carbon Flux from Land Cover Change. Houghton, R.A., 1991. Tropical deforestation and atmospheric carbon dioxide. In Tropical Forests and Climate. Springer, Netherlands, pp. 99–118. Houghton, R.A., 1999. The annual net flux of carbon to the atmosphere from changes in land use 1850–1990*. Tellus B. 51, 298–313. Houghton, R.A., 2008. Carbon flux to the atmosphere from land-use changes: 1850–2005. TRENDS: A Compendium of Data on Global Change. Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, U.S. Department of Energy, Oak Ridge, TN, U.S.A. IPCC, 2006. In: Eggleston, H.S., Buendia, L., Miwa, K., Ngara, T., Tanabe, K. (Eds.), Intergovernmental Panel on Climate Change. Guidelines for National Greenhouse Gas Inventories, Prepared by the National Greenhouse Gas Inventories Programme. IGES, Japan. Iverson, L.R., Brown, S., Prasad, A., Mitasova, H., Gillespie, A.J.R., Lugo, A.E., 1994. Use of GIS for estimating potential and actual forest biomass for continental South and Southeast Asia. In Effects of Land-Use Change on Atmospheric CO2 Concentrations. Springer, New York, pp. 67–116. Jennings, S.B., Brown, N.D., Sheil, D., 1999. Assessing forest canopies and understory illumination: canopy closure, canopy cover and other measures. Forestry 72 (1), 59–74. Jha, C.S., Fararoda, R., Rajashekar, G., Singh, S., Dadhwal, V.K., 2015. Spatial distribution of biomass in indian forests using spectral modelling. Multi-Scale Forest Biomass Assessment and Monitoring in the Hindu Kush Himalayan Region: A Geospatial Perspective. pp. p.139.

752