Materials and Method

Study Area

Karnataka covers an area of 191,976 km² (19 million hectares) with a share of 6% in the national GDP is located in the southern part of India, sharing borders with Maharashtra and Goa; Andhra Pradesh and Telangana to the east; Tamil Nadu and Kerala to the south, while the Arabian Sea forms the western boundary (Fig. 1). The population of the state was 611,30,704 inhabitants (as per 2011 census) with a density of 319 per km². The state is known for its diverse culture, scenic beauty, languages, economic, and social profiles. Karnataka state is divided into four revenue divisions, 49 subdivisions, 30 districts, 175 taluks, and 745 hoblies/revenue circles for decen- tralized administration. The Western Ghats, one of the 36 global biodiversity hotspots (https://www.conservation.org) covers 60% of the state’s forest cover in the western portion with diverse flora and fauna. The region has diverse forest cover types such

as evergreen, moist as well as dry deciduous, scrub, thorny, sholas, grasslands, and mangroves in the estuarine areas. The state harbors 4500 species of flowering plants, 508 species of birds, 150 varieties of mammals, 156 reptile species, amphibians of 156 species, 405 fish species, and 330 butterflies. Soils of the state are fertile by two major river systems (Krishna, Cauvery) and its tributaries. The state has a protected area network of five national parks (2431.3 km²) and 21 wildlife sanctuaries (3887.83 km²), covering nearly 16% of forest area. Agriculture and horticulture sectors are the backbone of the state’s economy. The state is the prime destination for IT and BT technologies with knowledge, innovation, research, and development centers. It has a gross domestic product (GDP) of |15.10 lakh crore (US$220 billion) as fourth largest in India, growing at a healthy 7% per year with a per capita GDP of |207,000 (US$3,000).

Method

The protocol adopted to assess the carbon dynamics in Karnataka is presented in Fig. 2. The research involved (i) assessment of land-use dynamics through spatial data acquired using spaceborne sensors at regular intervals; (ii) field data collection to classify remote sensing data, (iii) quantification of AGB through field measurements of girth and height and sampling of the locations through transect based quadrat;

(iv) quantification of carbon across various forest types and soil; (v) data mining pertaining to carbon emissions, sequestrations in forests and soils through published literature; (vi) visualization of likely changes in carbon dynamics (a) with the current rate of deforestation and degradation; (b) interventions with the afforestation; (c) implementation of the proposed development projects. This was implemented in three phases. Phase 1 focused on the land-use analyses, Phase 2 estimates the carbon sinks as well as its variation over time; quantified the emissions across each sector followed by carbon budgeting, and likely changes in carbon dynamics are predicted in Phase 3.

Land-use dynamics—Spatial patterns of land-use dynamics assessment using temporal remote sensing data: The remote sensing data of Landsat series for 1985, 2005, 2019 (downloaded from the public domain https://landsat.org) were analyzed through efficient supervised classifier based on GMLC (Gaussian Maximum Like- lihood Classifier) algorithm using free and opensource GRASS GIS (Geographical Analysis Support System—https://wgbis.ces.iisc.ernet.in/grass/). The field investi- gation has been carried out for collecting training data, which was used to classify the remote sensing data of 2019 coinciding with the field data collection period. The earlier time remote sensing data were classified using collateral data compiled from various sources such as Karnataka Forest Department reports (https://aranya.gov.in), vegetation map of South India of 1:250,000, the French Institute of India (https:// www.ifpindia.org). The process of remote sensing data classification involved (i) preparation of false-color composite (FCC) using five bands (R, G, and NIR) of LANDSAT satellite data, which assisted in the selection of training sites through the

identification of heterogeneous patches corresponding to diverse landscape elements,

(ii) attribute data collected in the field corresponding to these training polygons using precalibrated GPS (Global positioning system) and virtual data (Google Earth— https://earth.google.com, Bhuvan https://bhuvan.nrsc.gov.in), (iii) classification of RS data for eight different land-use categories through GMLC algorithm using training data, and (iv) accuracy assessment of classified remote sensing informa- tion was done through error matrix (contingency matrix) and K^Astatistics (Kappa). The training data compiled from field (60%) have been used for classification, while the balance is used for accuracy assessment and validation [28].

Estimation of Spatiotemporal Carbon Sequestration Potential

The carbon sequestration potential of forest ecosystems, plantations, and agricul- ture areas was assessed based on (i) field estimations carried out in the forests across Karnataka state using transect cum quadrat based sampling techniques and (ii)

published literature based on the rigorous distinctive biomass experiments. The study region (Karnataka State) was divided into 2597 grids of 5¹ 5¹ (or 9 km 9 km) grids corresponding to 5¹ 5¹ grids of 1:50,000 topographic maps of the Survey of India. Select grids corresponding to agroclimatic zones were chosen for biomass and carbon estimation through field investigations. The basal area, height, vegetation type (evergreen, deciduous, semievergreen, moist deciduous, scrub forests), diver- sity, biomass, carbon, etc., were estimated aiding field data. The comprehensive field evaluations were done across various forest cover types with about 424 transects in Uttara Kannada, Dharwad, Shimoga, Udupi, Chikmagalur, Dakshina Kannada, and Kodagu districts. The number of quadrats per transects varied between 3 and 5 depending on the occurrence of species in the sampling locality. The biomass was estimated using GBH or DBH (girth/diameter at breast height) for the trees >30 cm. The transect data and standard literature data were used for biomass quantifica- tion. The biomass, annual increment in biomass of various forest types, sequestered carbon, productivity have been computed using field data integrating with the infor- mation compiled from literature, which are listed in Tables 1 and 2. The probable relationship of biomass with the vegetation cover has been evaluated through multi- variate regression analysis across coastal, Sahyadri, plain regions of Karnataka. The carbon for above ground vegetation is computed as 50% of AGB value. The carbon is deposited in the soil as soil organic matter in both organic (SOC) and inorganic forms. SOC is calculated based on the field estimations in top 30 cm soil for different forests (Table 3) and mean soil carbon reported in the literature [31, 55, 40].

Quantification of Carbon Emission from Various Sectors

Data pertaining to emission for sectors such as agriculture, livestock, industry, energy, transportation, etc., were compiled from published literature.

Agriculture

Agricultural residue burning is practiced in some taluks across Karnataka. Emissions due to crop residue burning were computed as per the guidelines of IPCC and liter- ature [11, 61] based on the area of crop grown to the standard crop residue ratio. The total emission is estimated by summing of CO₂; methane (CH₄) and Carbon monoxide (CO) (its equivalent CO₂) values [17, 33]. The emission from agriculture residue (AR_e) burning is estimated as,

AR_e = Σ³_i=1 Σ[Cropresidueratio × emissioncoefficient]

Table 1 Forest type-wise quantification of biomass and sequestered carbon

Table 2 Above ground biomass for different forest types and plantations

Table 3 Soil carbon storage in different forest types and agriculture filed

Livestock

Livestock plays an important role in the agroecosystem, apart from the critical energy input to the croplands, also provides economic support to the farmers in terms of milk, manure, soil nutrient enrichment, etc. Livestock also produces CH₄ emissions from enteric fermentation and CH₄ and N₂O (nitrous oxide) emissions are from manure management systems. The agriculture sector accounts for approximately 20 and 35% of global GHG emissions [11]. The grid-wise livestock density has been estimated and associated emission was quantified under enteric fermentation as well as manure management [9, 21, 60]. Livestock population (Census 2012) data were obtained from the State Veterinary Department, Government of Karnataka, and respective emission factors are listed in Table 4. CH₄ emissions (kg CH₄/animal/year) due to the enteric fermentation are computed as,

Table 4 Emission factors associated with livestock

CH₄EntericFermentation = Σ_I(EF_I × N_I)/10⁶ (2)

where, EF_I is an emission factor for the individual livestock category, N_I is the number of animals of livestock for category I. The emission from manure depends on volatile solids or ruminants, their productivity, and manure handling system [51]. Methane emission due to manure management is estimated as,

CH₄Manuremanagement = Σ_I(EF_I × N_I)/10⁶ (3)

where EF_I is an emission factor for each livestock category; N_I is the number of livestock for category I in the region. Further, CO₂ equivalent values have been estimated across the grids for the fermentation and manure emissions.

Paddy cultivation is another major activity across the globe, contributing for 20% methane emission [64]. Paddy is grown in all taluks of Karnataka state and emission from paddy (Oryza sativa) is estimated across the grids as,

CH₄Paddy = [EF × T × A] (4)

where EF is the daily emission factor (kg CH₄/Ha/Day), T is the cultivation period, A is the harvested area (in two seasons-Kharif; Rabi).

Domestic

The fuelwood consumption is causing deforestation and an increase in C02 emis- sion. The Per Capita Fuel Consumption (PCFC) was analyzed to account fuelwood consumption pattern across the agroclimatic zones of state and determined the carbon emissions (EFC) as,

EFWC = [NH × PC FC × EF ] (4)

where EFWC is carbon emission from fuelwood consumption in rural households, PCFC is per capita fuel consumption (which was computed as ration of fuelwood consumed in kgs/day and number of adults in a household), EF is emission factor.

The waste generated per household level is also contributing to the CH₄ emissions due to the disorganized management of waste across the state. The waste generated across individual households of Karnataka at the grid level has been estimated consid- ering 0.35 g per person per day [45]. The average of four people per household was considered for a total of 2,281,419 households. The emission from waste per year is calculated as,

EWC = [0.35 × NH × 365] (5)

where EWC is carbon emission from waste generated, NH is the number of households.

Nitrous oxide (N₂O) emissions can occur as both direct and indirect emissions, apart from CH₄ through domestic wastewater, which has significant carbon loading [16]. The emission from wastewater generated from the individual households is estimated by considering average water consumption per person per day as 135 L [39, 45, 59].

Industries

Karnataka state is endowed with rich mineral resources as well as a large pool of human resources. The state has public sector units and also gives impetus simul- taneously to private sector growth, which prompted to establish many industries. The state has major manufacturing industries due to progressive industrial policies. The good Institutional networks such as Search Results Karnataka Industrial Area Development Board-KIADB (en.kiadb.in), Karnataka State Small Industries Devel- opment Corporation Ltd-KSSIDC (kssidc.co.in), Karnataka State Small Industries Development Corporation Ltd-KSSIMC, Technical Consultancy Services Organi- sation of Karnataka-TECSOK (tecsok.com), Federation of Karnataka Chambers of Commerce and Industry-FKCCI (fkcci.org), and Industries and Commerce Depart- ment (kum.karnatka.gov.in) were set up to provide various assistance for industrial development in the state. The major manufacturing industries such as cement, steel, iron ore processing, petrochemical, sugar, paper, and paper board, etc., were consid- ered [12] and associated emissions [51] were estimated based on the standard protocol (Annexure-I).

Energy

The energy sector is considered to account emissions from thermal (burning of coal) and diesel power generation. The state has an installed power generation capacity of 28,789.99 MW of which, central utilities contribute 4123 MW, private utilities contribute 13,259.71 MW and 11,407.28 MW under state utilities. The thermal power contributes 9,560.82 MW, 698.00 MW by nuclear, and 8,431.34 MW by renew- able energy sources for the total installed power generation capacity. The various thermal and diesel power generating units were mapped across the state and emissions associated were estimated (Annexure-II).

Transportation

Karnataka stands fifth as per registered motorized vehicles and contributing 7% of registered vehicles of India. Bengaluru has a large quantum of vehicles after Delhi with higher vehicle registrations. The quantum of registered vehicles in the state

has been gradually increasing at an average growth rate of 10% per annum and the decadal growth rate of vehicles at 138%. The two-wheelers account for 70% of the registered vehicles across the six divisions (Table 5). The emission from each type of vehicle was evaluated by computing annual average distance traveled (AADT) [15, 39, 44]. The total emission from the transportation sector has been quantified as,

Et = ΣV_i × AADT _i × EF_{i, jkm} (6)

where, E_t is the total emission from the transportation sector, V_i Number of vehicles per type i, AADT_i is the annual average distance traveled per different vehicle types and EF_i,j,km is the vehicle type (i) emission of factors (j), per driven kilometer (Table 6).

Carbon Ratio (CR)

CR was computed as a ratio of total carbon uptake (from AGB, SOC) to the total emissions across all sectors, which will provide the carbon status across the grids in Karnataka. CR values of “0” and close to 0 represent the regions of higher emission and value greater than 1 represent carbon sequestration is higher in that grids.

CR = Σ(Carbon Sequestration)/Σ(Carbon Emission)