Life Cycle Carbon Costs of Fibreboard, Pulp and Bioenergy Produced from Improved Oil Camellia (Camellia oleifera spp.) Forest Management Operations in China

Abstract

Oil camellia (Camellia oleifera) residues from low-yield forests offer significant potential for carbon emission reductions across multiple product pathways—fibreboard, pulp, and bioelectricity. Life cycle assessments (LCAs) were conducted for these three products, revealing distinct carbon footprints driven by energy use, chemical inputs, and combustion processes. Fibreboard production showed a carbon footprint of 244.314 kg CO₂e/m³, primarily due to diesel use and electricity consumption. Pulp production exhibited the highest carbon intensity at 481.626 kg CO₂e/t, largely driven by chemical consumption and fossil fuels. Bioelectricity, with the lowest carbon footprint of 41.750 g CO₂e/kWh, demonstrated sensitivity to transportation logistics and fuel types. Substitution and scenario analysis showed that emission reductions can be achieved by optimizing energy structure, substituting high-carbon chemicals, and improving transportation efficiency. The findings highlight the substantial reduction potential when oil camellia residues replace conventional feedstocks in these industries, contributing to the development of low-carbon strategies within the bioeconomy. These results also inform the design of targeted mitigation policies, enhancing carbon accounting frameworks and aligning with China’s dual-carbon goals.

1. Introduction

The escalating imperative for climate change mitigation has placed forestry-based resources at the forefront of global bioeconomy strategies, particularly emphasizing the valorization of underutilized biomass for low-carbon material and energy substitution [1]. In China, oil camellia (OC, Camellia oleifera spp.), a perennial woody oil crop, spans over 11 million hectares across subtropical regions [2,3,4]. However, more than 70% of these plantations are classified as low-yield forests, generating substantial volumes of residual biomass—including pruned branches, bark, fruit shells, and foliage—during forest renovation and oil harvesting operations [5,6,7]. Despite their abundance and high lignocellulosic content, these residues are often discarded or openly burned, resulting in considerable resource underutilization and unaccounted greenhouse gas (GHG) emissions [8,9]. National policy directives, such as the 14th Five-Year Plan for Bioeconomy Development and the Action Plan for Carbon Dioxide Peaking Before 2030, have underscored the importance of integrating forestry residues into circular material systems [10,11]. Yet, the environmental implications of converting oil camellia biomass into industrial products remain poorly quantified.

Life cycle assessment (LCA) has emerged as the principal methodological framework for evaluating the environmental performance of bio-based products [12]. However, the existing literature reveals several critical limitations that undermine the applicability of LCA to OC-derived systems [13,14]. A primary shortcoming is the persistent material focus bias. Most LCA studies on wood-based panels and pulp concentrate on conventional commercial species such as eucalyptus, pine, or bamboo, yielding carbon footprint values ranging from 357 to 759 kg CO₂e/m³ for fibreboard and 388 to 722 kg CO₂e/t for pulp depending on processing techniques [15,16,17,18]. By contrast, OC residues remain largely unstudied despite notable differences in chemical composition, processing compatibility, and regional logistics. Furthermore, most LCAs are limited to single-product systems, analyzing either wood panels or biomass energy in isolation [19]. This narrow scope precludes a holistic understanding of emissions trade-offs and conversion synergies across multiple pathways derived from a common feedstock—a key oversight in the context of OC residue utilization [19,20]. Methodological inconsistencies further compound these limitations. Although ISO 14040/44-compliant frameworks are widely adopted, the definition of system boundaries in many LCA studies remains overly narrow, with a predominant focus on cradle-to-grave scopes [21]. This constrained boundary setting systematically excludes downstream processes such as product use, end-of-life treatment, and potential substitution effects—elements that are particularly critical when evaluating the full climate mitigation potential of bio-based systems [19]. Such methodological truncation not only limits the comprehensiveness of emission accounting but also underrepresents the broader environmental value of cascading biomass utilization [22]. Recent high-impact reviews have highlighted this issue as a pervasive shortcoming: Bortoli et al. (2025) [23] report that the majority of bioenergy-related LCAs are confined to upstream stages, while Hosseinzadeh-Bandbafha et al. (2021) find that fewer than 17% of published assessments extend to cradle-to-grave boundaries, underscoring a systemic underestimation of avoided emissions and long-term carbon storage opportunities [24,25].

Despite the growing relevance of LCA for carbon labeling and crediting schemes, many studies fail to contrast bio-based products against conventional alternatives or assess their potential contribution to mitigation frameworks such as the China Certified Emission Reduction (CCER) program or Article 6 of the Paris Agreement [26]. Moreover, the literature reviews in this domain often remain descriptive, offering limited problematization or conceptual synthesis. Key research gaps—such as the systemic behavior of marginal biomass in multi-pathway conversion, or the functional trade-offs between short-lived energy outputs and long-term material storage—are rarely identified, let alone hypothesized [19,27].

This study seeks to address these deficiencies through a process-based, cradle-to-grave life cycle carbon accounting of three distinct products—fibreboard (per m³), pulp (per t), and bioelectricity (per kWh)—produced from oil camellia renovation residues. The central hypothesis posits that although these product systems share a common biomass origin, they exhibit fundamentally divergent emission profiles, process drivers, and mitigation potentials. We further argue that prevailing generalizations across product systems obscure critical inter-pathway differences, thereby diminishing the resolution and policy relevance of LCA as a decision-support tool [28]. Drawing on primary inventory data collected from representative industrial facilities in Guangxi province, and utilizing regionally calibrated emission factors, we construct an integrated LCA model that captures material input, energy consumption, transportation, processing, and auxiliary inputs. Beyond baseline quantification of global warming potential (GWP), the study evaluates substitution scenarios, quantifying the avoided emissions when OC-derived products replace fossil-intensive alternatives.

The novelty of this research lies in its integrative modeling of multi-output systems derived from forestry renovation residues, its disaggregated attribution of emission drivers grounded in empirical data, and its linkage of carbon outcomes to policy-relevant substitution mechanisms. By advancing both methodological rigor and contextual specificity, this study contributes not only to the empirical knowledge base surrounding forestry-based carbon accounting, but also to the broader discourse on climate-compatible resource transitions and sustainable biomass utilization.

2. Materials and Methods

2.1. System Definition

In LCA, it is essential to clearly define system boundaries, including the interface between the technological system and the natural environment, the geographic and temporal scope, and the complete life cycle of the product under investigation [29]. This study focuses on the valorization of transformation residues from low-yield oil camellia (OCLYF) forests to produce three primary product categories—fibreboard, pulp, and bioenergy—within the context of three forest management strategies: forest tending, crown replacement, and reforestation. The overall LCA framework and major material flows associated with each pathway are illustrated in Figure 1. The system boundaries and major material flows for each pathway are illustrated in Figures S1–S3 (Supplementary Material), corresponding to the lumber production, pulp production, and bioenergy production pathways, respectively.

The systems evaluated comprise (a) fibreboard production (Figure 1a), (b) pulp production (Figure 1b), and (c) biomass energy generation (Figure 1c). Residual materials, including logs, branches, leaves, taproots, and lateral roots, are sorted and allocated to the corresponding downstream systems. Log residues are first transported to lumber yards for peeling and pretreatment, during which bark residues are redirected to the biomass energy system [30]. The peeled logs are subsequently processed in wood manufacturing facilities, where additional by-products such as wood chips and ground material are also diverted for bioenergy use [31]. The remaining solid wood is refined into fibreboard, while smaller wood fragments are transferred to the pulp production system [32]. Final fibreboard and pulp products are transported to end users. Concurrently, thermal energy generated in the biomass system is either consumed onsite or converted into electricity for internal or external use.

Figure 1. Main processes and inputs accounted for during primary, secondary, and end-of-use stages within the consequential life cycle assessment of oil camellia (OC, Camellia oleifera) products (Adapted from Ref [33]). (a) The low-yield forest provides logs and residues. (b) The lumber production system converts logs into lumber and residues. (c) The pulp mill processes logs into pulp, generating by-products. (d) The bioenergy system uses residues for combined heat and power generation. (e) The end-of-use phase includes disposal, recycling, and substitution of fertilizer with combustion ash. See also detailed system boundaries in Figures S1–S3.

Figure 1. Main processes and inputs accounted for during primary, secondary, and end-of-use stages within the consequential life cycle assessment of oil camellia (OC, Camellia oleifera) products (Adapted from Ref [33]). (a) The low-yield forest provides logs and residues. (b) The lumber production system converts logs into lumber and residues. (c) The pulp mill processes logs into pulp, generating by-products. (d) The bioenergy system uses residues for combined heat and power generation. (e) The end-of-use phase includes disposal, recycling, and substitution of fertilizer with combustion ash. See also detailed system boundaries in Figures S1–S3.

2.2. Goal and Scope, Functional Units

This study aims to evaluate the cradle-to-grave carbon footprint (CF) of three product systems—fibreboard, pulp, and biomass energy—using LCA methodology. Specifically, the analysis quantifies emissions from raw material acquisition, transportation, and processing phases [32]. The defined functional units are as follows: 1 cubic meter (m³) of OC based fibreboard product conforming to EN 312 standards, 1 tonne (t) of OC based pulp product, and 1 kilowatt-hour (kWh) of bioelectricity generated from OC biomass.

2.3. System Boundaries

The system boundary for this study follows a cradle-to-grave approach, encompassing all major life cycle stages from raw material acquisition to final disposal (Figure 1 and Figures S1–S3). As illustrated in Figure S1, the analysis includes upstream biomass harvesting, on-site and off-site transportation, industrial processing, product use, and end-of-life treatment. This boundary configuration enables a comprehensive evaluation of the carbon implications associated with each product pathway.

The cradle-to-grave stages cover the extraction of OC residues, subsequent transportation to processing facilities, and the conversion into three product forms: fibreboard, pulp, and bioelectricity. Inputs to these stages include biomass feedstocks, thermal and electrical energy, chemical additives, auxiliary materials, and infrastructure use. Outputs primarily consist of greenhouse gas emissions (CO₂, CH₄, N₂O) generated during fuel combustion, chemical reactions, and material handling. The use-phase is incorporated for material-based products. Fibreboard is modeled as a long-lived product with negligible in-use emissions during its functional period, while pulp products are characterized by short service lives and rapid entry into post-consumer waste flows. The bioelectricity pathway does not involve a use-phase, as energy is consumed at the point of generation.

End-of-life processes are considered for fibreboard and pulp pathways and include three main treatment scenarios: landfill, incineration with energy recovery, and material recycling. Allocation of post-consumer products into each disposal route is based on nationally representative data. Emissions associated with these processes are calculated using scenario-specific activity data and emission factors, as detailed in Section 2.7.6. Figures S2 and S3 delineate the system boundaries specific to the pulp and bioenergy pathways, respectively, each incorporating upstream, processing, and downstream stages in alignment with the unified cradle-to-grave framework.

2.4. Data Collection

Primary data were obtained through field surveys and on-site visits to the Guangxi Weidu State-owned Forest Farm. The information regarding the labor requirements for road construction and the transformation of low-yield forests, as well as other input factors, was collected through structured interviews with local forestry personnel (including Cao Chunrui, the manager of the plantation).

Given the limited availability of region-specific data for the processing of OC residues into fibreboard, pulp, and bioenergy, the life cycle inventory (LCI) was supplemented with secondary data sources. Energy consumption was estimated based on established parameters from prior assessments [34], while data related to harvesting operations and wood processing adhered to standardized inventory frameworks developed for forestry-based materials [34,35]. Pulp production processes were informed by empirical data from wood-based pulp mills [36], and emission coefficients associated with labor and forestry infrastructure were extracted from nationally reported field-level inventories [36,37].

Energy use and emissions associated with transportation were calculated based on the official “Greenhouse Gas Emission Accounting Methods and Reporting Guidelines for Land Transportation Enterprises” [38]. Machinery-related emissions were referenced from the “Greenhouse Gas Emission Accounting Guidelines for Facility Agricultural Enterprises” [39] (see Table S1). Additional emission factors for electricity, fuels, and materials were drawn from authoritative databases including CityGHG, ELCD, Ecoinvent 3.0, The Climate Registry (TCR), and Industry Data 2.0 [38,40,41,42]. Transportation distances for both road and rail were determined using the China Land Transportation Network (Table 1).

Table 1. Data related to forest road building and timber production and their sources.

Data Components	Data Sources and References *
Forest road construction data	Weidu forest farm field data
Low modification of labor input data	Weidu forest farm field data
On-site transportation methods and distances	Weidu forest farm field data
Energy consumption data for harvesting	[43]
Energy and material consumption data for lumber production	[35]
Energy and material consumption data for pulp production	[36]
Energy consumption data for biomass production	[35]
Labor use	[37]
Emission coefficients for trail facilities	[41,42,44]
Transportation and energy consumption data	[38]
GHG emission coefficients for transportation fuels	[38]
GHG emission coefficients for forestry machinery use	[39]
Emission coefficients of upstream products, electricity, fuel oil, natural gas and coal	[16,40,41,42,45]
Road and rail transportation distance	[40]

Note: * The life cycle inventory data were compiled from multiple sources, including field-collected records for forest operations and transportation, literature-derived values for energy and material inputs, and standardized emission factors based on official guidelines. All datasets correspond to unit processes defined within a gate-to-gate system boundary and are applicable to lumber production, pulp manufacturing, and biomass energy pathways.

Table 1. Data related to forest road building and timber production and their sources.

Data Components	Data Sources and References *
Forest road construction data	Weidu forest farm field data
Low modification of labor input data	Weidu forest farm field data
On-site transportation methods and distances	Weidu forest farm field data
Energy consumption data for harvesting	[43]
Energy and material consumption data for lumber production	[35]
Energy and material consumption data for pulp production	[36]
Energy consumption data for biomass production	[35]
Labor use	[37]
Emission coefficients for trail facilities	[41,42,44]
Transportation and energy consumption data	[38]
GHG emission coefficients for transportation fuels	[38]
GHG emission coefficients for forestry machinery use	[39]
Emission coefficients of upstream products, electricity, fuel oil, natural gas and coal	[16,40,41,42,45]
Road and rail transportation distance	[40]

Note: * The life cycle inventory data were compiled from multiple sources, including field-collected records for forest operations and transportation, literature-derived values for energy and material inputs, and standardized emission factors based on official guidelines. All datasets correspond to unit processes defined within a gate-to-gate system boundary and are applicable to lumber production, pulp manufacturing, and biomass energy pathways.

2.5. Feedstocks and by Products

During the transformation of OC residues into finished products, each processing pathway generates distinct process-derived by-products that may influence material efficiency and environmental performance. These by-products were carefully inventoried and, where relevant, included in the system modeling according to ISO 14044 standards for multi-output systems [46].

In the fibreboard production system, typical by-products include sanding dust, panel trimmings, and oversized particles rejected during fiber screening [47]. Sanding dust, rich in fine lignin-cellulose particles, is frequently recovered as a fuel substitute for on-site thermal energy, contributing to internal energy loops [48]. Trimming waste is either redirected to low-grade filler production or sold to local small-scale combustion users where applicable. Oversize rejects are typically reprocessed or combusted, depending on fiber integrity [47]. All recyclable fractions are accounted for through internal material recycling credits, whereas non-recyclable losses are modeled as solid waste outputs with corresponding transport and disposal emissions.

In the pulping system, the dominant by-product is black liquor, a lignin-rich effluent from the chemical digestion stage [49]. A significant portion is recovered via chemical recovery boilers and reused for process steam generation, representing a key source of internal energy. However, unrecovered fractions (due to system loss or capacity limits) are modeled as wastewater discharges with dissolved organics, contributing to potential waterborne emissions [50]. In addition, screening rejects and short fibers are separated before drying and are either combusted or discarded, depending on mill configuration. These outputs are treated as non-core flows unless reused or sold as low-grade feedstocks.

In the bioelectricity pathway, combustion produces two main residuals: bottom ash and fly ash [51]. Bottom ash, primarily consisting of non-volatile inorganic minerals, is removed from the furnace bed and typically disposed of in landfills or used as construction fill where allowed. Fly ash, captured via electrostatic precipitators or bag filters, may be classified as hazardous depending on heavy metal content. Where regulatory conditions permit, ash residues are used as soil amendments or cement additives; otherwise, they are treated as controlled waste and modeled accordingly [49,52].

2.6. Process Allocation

This study adopts a mass-based allocation method to distribute environmental burdens—including energy consumption, material inputs, transportation, and process outputs—across the different co-products generated from OC transformation residues (Table S1). In the fibreboard production pathway, representative inputs include lubricants used in logging operations, hydraulic fluids for mechanical systems, and solvent-based chemicals for surface treatments. The pulp production process incorporates a broader set of chemical reagents, including calcium oxide and sodium hydroxide for pulping, and hydrogen peroxide, sulfuric acid, ethanol, and sodium chlorate for bleaching stages.

Energy inputs vary substantially across the three systems. For fibreboard manufacturing, energy sources comprise diesel (used during harvesting), propane (for thermal processing), grid electricity, wood boiler-generated electricity, natural gas (for drying operations), and gasoline (for mobile equipment). In contrast, pulp production is powered by a combination of grid and on-site electricity, steam generated from internal sources, and coal combustion. Biomass energy production systems rely on natural gas and liquefied petroleum gas as primary fuels to support combustion and cogeneration processes.

Transportation logistics were structured according to the specific requirements of each production pathway. In the fibreboard system, logging residues are first transported within the forest over an average distance of 5.3 km using diesel-powered pickup trucks. This is followed by 80 km of regional haulage via 8 t medium-duty trucks to the nearest rail terminal. Materials are then conveyed approximately 285 km by freight rail, before undergoing a final 45 km of delivery to the processing facility using gasoline-fueled vehicles. Similar multi-modal transport configurations are applied in the pulp and bioenergy systems, with adjustments made to transport distances and vehicle specifications based on site-level conditions. All emission factors associated with transportation were sourced from nationally recognized guidelines and official carbon accounting databases.

The allocation of transformation residues across the three product pathways was determined based on dry mass ratios, accounting for differences in material density and moisture content inherent to OC wood components. This approach ensures a consistent and physically representative partitioning of environmental loads among co-generated outputs.

2.7. Carbon Footprint Modeling

2.7.1. Material Use

The emissions from material use, including chemical inputs such as paint solvents, calcium oxide, sodium hydroxide, sodium chlorate, hydrogen peroxide, ethanol, and sulfuric acid throughout the product life cycle, as well as lubricating oil, hydraulic oil, and water used in production, are accounted for. The corresponding GHG emissions are calculated as follows:

$$C_1=\sum _i^n P_i\times E{F}_i$$

(1)

where C₁ represents the GHG emissions from materials used to produce wood products from OCLYF residues, P_i is the material category, is the total material used, and EF_i is the emission factor for category i; i denotes the i-th material category. The material-related emission factors were as follows: H₂O₂ 2.316 kg CO₂e/kg [38], NaOH 1.12 kg CO₂e/kg [38], NaClO₃ 0.8638 kg CO₂e/kg [38], H₂SO₄ 0.0776 kg CO₂e/kg [40], water 0.213 kg CO₂e/m³ [40], CaO 1.11 kg CO₂e/kg [41], paint solvent 1.19972 kg CO₂e/kg [42], hydraulic oil 0.20 kg CO₂e/L [53], lubricating oil 1.20 kg CO₂e/kg [53], and ethanol 1.63 kg CO₂e/L [44]. In practice, each consumption record from the life cycle inventory was matched to its corresponding emission factor, converted to CO₂ equivalents, and aggregated across all materials to yield the total material-related emissions for the given production pathway. The coefficients and sources are in Table S2.

2.7.2. Energy Use

Energy use emissions include grid and on-site electricity, diesel, propane, natural gas, coal, and liquefied petroleum gas consumed throughout the product life cycle. The GHG emissions are calculated as follows:

$$C_2=\sum _i^n P_i\times E{F}_i$$

(2)

where C₂ represents emissions from energy used in producing wood products from OC residues, i is the energy type, P_i is the energy consumed, and EF_i is the emission coefficient. Refer to Table S3 for coefficients and sources. The emission factors applied for energy use account for both direct combustion emissions and upstream emissions from extraction, processing, and delivery to the point of use. These factors include grid electricity 0.5207 t CO₂e/MWh [38], diesel fuel 3.797 t CO₂e/t [38], natural gas 1.072 kg CO₂e/kg [38], coal 0.08 kg CO₂e/kg [38], gasoline 2.3 kg CO₂e/L [53], and liquefied petroleum gas 5.68 kg CO₂e/gallon [43]. Consumption data from the life cycle inventory were matched to these factors, converted to CO₂ equivalents, and aggregated to obtain the total energy-related emissions within this stage of the assessment.

2.7.3. Transportation

Transportation emissions include those from road transportation of raw materials, on-site transportation within LYF land, and off-site transport of residual materials. The calculation is as follows:

$$C_3=\sum _i^n P_i\times M_i\times D_i\times E{F}_i$$

(3)

where C₃ represents the GHG emissions produced by the transportation of wood products derived from OCLYF transformation residues, i denotes the type of energy used in transportation, Pᵢ is the energy consumed, Mᵢ is the payload, Dᵢ is the transport distance, and EFᵢ is the corresponding emission factor. Refer to Table S4 for coefficients, including 2.71 kg CO₂e/t·km for diesel 8t trucks and 0.52 kg CO₂e/m for in-forest diesel pickup trucks, based on the China Product Carbon Emission Factor Database [39].

2.7.4. Labor Use

Due to the challenging terrain and heterogeneous distribution of LYFs, significant labor is required for logging and on-site transportation. The manual GHG emission calculation is as follows:

$$C_4=P_i\times E{F}_{labor}$$

(4)

where C₄ represents emissions from labor in producing wood products from OC residues, Pi is labor input in man-days, and EF_labor is the emission coefficient, 0.86 kg CO₂eq/person-day [37].

2.7.5. Infrastructure

Infrastructure emissions are from the construction, operation, and maintenance of infrastructure, primarily forest road construction and maintenance. The calculation is as follows:

$$C_5=L\times E{F}_{load}$$

(5)

where C₅ represents emissions from forest road maintenance, L is the activity level expressed in kilowatt-hours per cubic meter, and EF_load is the emission factor for load maintenance, estimated at 0.5207 kg CO₂eq/kWh based on national carbon accounting guidelines [39].

2.7.6. End-of-Use

In the fibreboard production system, typical by-products include sanding dust, panel trimmings, and oversized particles rejected during fiber screening. Sanding dust, rich in fine lignin-cellulose particles, is frequently recovered as a fuel substitute for on-site thermal energy, contributing to internal energy loops. Trimming waste is either redirected to low-grade filler production or sold to local small-scale combustion users where applicable. Oversize rejects are typically reprocessed or combusted, depending on fiber integrity. All recyclable fractions are accounted for through internal material recycling credits, whereas non-recyclable losses are modeled as solid waste outputs with corresponding transport and disposal emissions.

In the pulping system, the dominant by-product is black liquor, a lignin-rich effluent from the chemical digestion stage. A significant portion is recovered via chemical recovery boilers and reused for process steam generation, representing a key source of internal energy. However, unrecovered fractions (due to system loss or capacity limits) are modeled as wastewater discharges with dissolved organics, contributing to potential waterborne emissions. In addition, screening rejects and short fibers are separated before drying and are either combusted or discarded, depending on mill configuration. These outputs are treated as non-core flows unless reused or sold as low-grade feedstocks.

In the bioelectricity pathway, combustion produces two main residuals: bottom ash and fly ash. Bottom ash, primarily consisting of non-volatile inorganic minerals, is removed from the furnace bed and typically disposed of in landfills or used as construction fill where allowed. Fly ash, captured via electrostatic precipitators or bag filters, may be classified as hazardous depending on heavy metal content. Where regulatory conditions permit, ash residues are used as soil amendments or cement additives; otherwise, they are treated as controlled waste and modeled accordingly.

All process by-products were assigned to elementary or techno-specific flows depending on their fate. Emissions and environmental credits were allocated based on mass, energy, or economic value, as described in Section 2.5. Flows not entering the techno sphere or having negligible environmental relevance were excluded under cut-off criteria per ISO guidance.

2.7.7. Total Carbon Footprint

The total CF for the life cycle of board, pulp, and biomass energy products from OC residues is the sum of emissions from materials, energy, transportation, labor, and infrastructure, calculated as follows:

$$C=\sum _i^n C_i$$

(6)

where C is the total CF (kg CO₂eq) of products from OC residues, i is the type of carbon emissions, and C_i is the amount of each type of emission (kg CO₂eq).

2.8. Environmental Impact Assessment

Environmental impacts were quantified based on a life cycle impact assessment (LCIA) framework in accordance with the ISO 14044 standard and guided by the IPCC methodology for greenhouse gas inventories [46]. A unified characterization approach was adopted to compute each impact category using the following generalized formula:

$${EI}_c=\sum _i(A_i\times {EF}_{i,c})$$

(7)

where EI_c is the total environmental impact for category c, A_i represents the activity data (e.g., fuel consumption, emissions of pollutants, or material inputs), and EF_i,c is the category-specific emission or impact factor corresponding to flow iii. For climate change impacts, IPCC’s 100-year global warming potentials from the Fifth Assessment Report (AR5) were used as characterization factors [54].

In the context of this study, activity data Ai were derived from the life cycle inventory (Section 2.4) of three OC—based systems: fibreboard (m³), air-dried pulp (t), and biomass electricity (kWh). The GWP indicator accounted for fossil CO₂, CH₄, and N₂O emitted across harvesting, processing, and transportation stages, with the respective factors of 1, 28, and 265 (kg CO₂-eq per kg gas) used for conversion.

2.9. Sensitivity Analysis

To assess the robustness of the environmental impact outcomes and quantify uncertainty propagation within the life cycle models, a combined local and global sensitivity analysis was performed. Local sensitivity was evaluated through ±20% deterministic perturbations applied to key inventory parameters, including emission factors, chemical input quantities, energy consumption levels, transportation distances, and allocation ratios, while maintaining all other variables constant to isolate the influence of individual parameters on selected impact categories [55,56]. For global sensitivity and probabilistic uncertainty assessment, Monte Carlo simulation was implemented using Oracle Crystal Ball 11.1. Assumed probability distributions for key parameters were based on data availability and variability characteristics, typically adopting triangular distributions for material inputs and lognormal distributions for emission factors and energy intensities, with 1000 iterations per scenario to ensure convergence. This simulation approach allowed for the characterization of cumulative uncertainty within each product system’s inventory, focusing on impact categories such as global warming potential (GWP).

3. Results

3.1. Life Cycle Inventory

The LCI of the OC resource cascade covers three distinct transformation pathways—fibreboard production, pulp manufacturing, and biomass-based power generation (Table 1). Each pathway operates under a cradle-to-grave boundary and reflects a different technological logic, resource intensity, and infrastructure dependency, ultimately shaping their environmental burdens and mitigation potential.

Fibreboard production represents the most input-diverse and process-dense system among the three (

Table 2. Life cycle inventory of lumber production derived from improved oil camellia OC forest management operations, from “harvesting gate to grave”.

Category	Item	Lumber		Pulp Production		Bioenergy
		Input Value	Unit	Input Value	Unit	Input Value	Unit
Energy	Electricity	54.087	kWh/m3	11.382	kWh/t	0.001	m/kWh
	Diesel	23.353	L/m3	4.666	kg/t	0.004	kg/kWh
	Gasoline	2.033	L/m3	0.443	L/m3	0.0316	t·km/kWh
	Propane	0.047	kg/m3			0.1	t·km/kWh
	Natural gas	4.235	m3/m3	7.859	GJ/t	0.0079	kg/kWh
	Onsite power	29.282	kWh/m3	56.413	kWh/t
	Coal			196.771	kg/t
	Onsite steam			7.859	GJ/t
Materials	Paint			0.000
	Water	274.670	kg/m3	27.563	m3/t	0.2	m3/m3
	Lubricant	0.014	kg/m3			0.0013	L/m3
	Hydraulic fluid	0.018	kg/m3
	Calcium Oxide			91.464	kg/t
	Ethanol			15.378	kg/t
	Hydrogen peroxide			3.595	kg/t
	Sulfuric acid			19.828	kg/t
	Sodium chlorate			29.098	kg/t
	Sodium hydroxide	0.014	kg/m3	36.761	kg/t	0.0094	person·day/kWh
Road	Road maintenance					0.0010	m/kWh
Labor	Labor	0.063	person·day/m3			0.0094	person·day/kWh
Transport	Gasoline truck	0.052	L/m3	0.103	t·km/kWh	0.0316	t·km/kWh
	Diesel 8 t truck	2.033	L/m3	4.666	kg/t	0.2	t·km/kWh
	Diesel pick uptruck	0.052	L/m3	0.044	L/m3
	Train			0.206	t·km/kWh	0.0316	t·km/kWh

Note: The life cycle inventory (LCI) covers the production of lumber, pulp, and bioenergy from OC residues. It includes inputs like energy (electricity, diesel, gasoline, natural gas, etc.), materials (water, lubricants, chemicals), labor (person·day), and transportation emissions (from trucks and trains). Road maintenance is also accounted for.

). It begins with plantation-stage harvesting, where 3.10 L of diesel and 0.048 L of lubricant per m³ of final product are used for mechanized felling and forwarding, supplemented by 3.00 kWh of national grid electricity for load maintenance and 0.052 person·day/m³ of manual labor. Transportation of logs combines rail-based haulage (2.30 L/m³) with minor truck use, optimized for centralized mill supply chains. The fibreboard mill itself is marked by exceptionally high water demand (>260 kg/m³), drawn from both surface and groundwater, required for material conditioning and cleaning.

A unique aspect of this pathway is its high energy diversity: grid electricity (54.087 kWh/m³), on-site wood boiler power (29.282 kWh/m³), and multiple fossil fuels—diesel (23.353 L/m³), gasoline (2.033 L/m³), natural gas (4.235 m³/m³), and liquefied petroleum gas (8.096 L/m³)—supporting cutting, pressing, drying, and finishing stages. Surface coating processes use solvent-based paints (0.001 kg/m³) and hydraulic fluid (0.018 kg/m³), while emissions include methanol (0.0204 kg/m³), formaldehyde, and acrolein, pointing to complex interactions between chemical formulation and thermal conditions. Dust (PM10 and PM2.5), VOCs from pressing, and wood ash residues (e.g., 2.299 kg/m³ boiler fly ash) add to the air and solid emission profile. Altogether, the fibreboard system is shaped by mechanical refinement, surface engineering, and multi-source energy reliance, revealing a tightly interwoven network of material transformation and emission pathways.

Pulp production, in contrast, follows a chemically intensive regime driven by large-scale digestion and bleaching. For each tonne of market-ready pulp, 2.75 m³ of OC wood is processed. The pulping line begins with 91.464 kg of CaO/t for fiber separation and pH adjustment, followed by oxidative agents including NaOH (36.761 kg/t fresh + 102.732 kg/t recovered), NaClO₃ (29.098 kg/t), H₂O₂ (15.378 kg/t), and CH₃OH (3.595 kg/t). This creates a strongly alkaline environment with high chemical throughput and effluent generation.

The energy matrix is similarly expansive: coal (196.771 kg/t), bunker oil (13.266 kg/t), diesel (4.666 kg/t), and internal steam use (7.859 GJ/t) are supported by electricity input (56.413 kWh/t), most of which is sourced from on-site cogeneration units. While the use of recovered NaOH and steam loops indicates partial circularity, the high input intensity and effluent volumes—effluent water (35.715 m³/t), TSS (0.618 kg/t), COD/BOD, NO_x, TRS, and ash (57.608 kg/t)—reflect a system geared toward throughput maximization rather than material efficiency. The process is further complicated by waste handling challenges and flue gas emissions, embedding environmental hotspots within its integrated chemical-energy-water triad.

Bioelectricity generation emerges as the leanest yet most logistically sensitive pathway. Utilizing non-commercial biomass fractions (branches, leaves), this system bypasses material refinement entirely. Its field operations consume just 0.0010 m/kWh of electricity for equipment loading and 0.00935 person·day/kWh of labor. Transport relies on rail (0.0316 t·km/kWh) and trucks (0.1000 t·km/kWh), but lacks any intermediate preprocessing. Combustion stabilization is achieved via liquefied petroleum gas (0.00395 kg/kWh) and natural gas (0.00790 kg/kWh), with no additional chemical inputs or waste streams recorded.

This direct-use configuration minimizes emissions and infrastructure dependency, but may underutilize the carbon sequestration potential embedded in residues. It represents a strategic compromise—low transformation cost versus limited value-added functionality—especially when deployed in decentralized energy contexts.

Across the three systems, the LCI reflects divergent technological priorities: fibreboard seeks material durability through high-intensity refinement; pulp demands chemical efficiency and volume; biomass energy prioritizes immediacy and flow-through combustion. These structural contrasts inform the pathways’ subsequent carbon performance and suggest tailored optimization levers—energy restructuring in fibreboard, chemical substitution in pulp, and routing efficiency in biomass systems.

3.2. Product Carbon Footprint and Supply Chain Emission

The product carbon footprint and supply chain emissions of OC-derived products were analyzed across three distinct pathways: fibreboard, pulp, and bioelectricity, as detailed in

Table 3. Greenhouse gas emissions from plantation-based production of OC wood fiber materials (FU = 1 m³ OC wood fiber materials).

Life Cycle	Consumption	GWP, kg CO2e/m3	%Total
Harvest and internal transport
	Diesel for ground-based harvesting	8.153	68.970
	Lubricant for harvesting	0.058	0.487
	Diesel for internal transport	0.870	7.361
	Lubricant for internal transport	0.002	0.013
	Load maintenance (electricity from national grid)	1.562	13.215
	Labor for harvesting	0.045	0.378
Transport (gate to gate)
	Transport by train (gasoline use)	0.023	0.195
	Transport by 8 t-truck (gasoline use)	0.117	0.987
Direct emissions
	Emissions to air 1	0.058	0.491
	Emissions to water 2	0.934	7.903
Total		11.821	100

Note: ¹ Air emissions include TN loss, CH₄, and NOₓ-N, calculated using IPCC (2006) [57] and Zhang et al. (2020) [52]. ² Water emissions include COD, BOD, TSS, and NH₃, estimated following Wang et al. (2025) [13] and IPCC guidelines, using simplified conversion factors (e.g., COD/BOD ≈ 1.5, 1 kg TSS ≈ 1.6 kg CO₂e). Emission factors are derived from the China Product Carbon Emission Factor Database (2022) [38], BEIS (2022) [58], and relevant literature sources as specified in the LCI.

,

Table 4. Greenhouse gas emissions from fibreboard production using OC wood (FU = 1 m³ chemical treated planed dry OC lumber).

Consumption	Inputs	GWP, kg CO2e/m3	%Total
Wood fiber materials
	OC wood fiber materials	17.307	7.089
Energy and fossil fuels
	Diesel	74.316	30.441
	Electricity from national grid	34.077	13.958
	Gasoline	6.102	2.499
	Liquefied petroleum gas	12.145	4.975
	Lube oil	0.020	0.008
	Natural gas	9.835	4.028
	On-site generated electricity	8.149	3.338
	Propane	0.086	0.035
Chemicals	Solvent paint, hydraulic fluid	0.006	0.003
Other materials	Water	81.331	33.314
Direct emissions 1		0.939	0.385
Total		244.134	100

Note: ¹ Direct emissions include saw dust, VOCs from drying and pressing, particulates, methanol, formaldehyde, acrolein, acetaldehyde, 1-Propanol, suspended solids, unspecified oils and combustion residues (boiler fly ash, pins).

,

Table 5. Greenhouse gas emissions by input/output category for market OC pulp production (FU = 1 t air-dried OC pulp).

Consumption	Items	GWP, kg CO2e/m3	%Total
Wood fiber materials
	OC pulpwood	32.509	6.750
Energy and fossil fuels
	Bunker oil	24.158	5.016
	Coal	15.742	3.268
	Diesel	17.716	3.678
	Electricity from national grid	5.926	1.230
	On-site generated electricity	29.374	6.099
	Internal used steam	1.305	0.271
Chemicals
	CaO	101.525	21.080
	CH3OH	5.874	1.220
	H2O2	35.615	7.395
	H2SO4	1.539	0.319
	NaClO3	25.135	5.219
	NaOH	41.246	8.564
	On-site recovered NaOH	20.546	4.266
Other materials
	Water	5.871	1.219
Direct emissions to air
	NOx	3.384	0.703
	SO2	2.331	0.484
	Particulates	0.005	0.001
	TRS	0.104	0.022
	Waste heat	28.165	5.848
Direct emissions to water
	COD	8.158	1.694
	BOD	1.579	0.331
	Total N	0.353	0.073
	Total P	1.255	0.260
	TSS	0.989	0.205
	Effluent	40.451	8.399
Solid wastes
	Ash and dust	4.609	0.957
	Sludge	26.146	5.429
Total		481.626	100

and

Table 6. Greenhouse gas emissions associated with the production of OC bioelectricity (FU = 1 kWh bioelectricity generated from OC feedstock).

Consumption	Inputs	GWP, g CO2e/kWh	%Total
Load maintenance	Electricity from national grid	0.536	1.285
Labor for harvesting	Labor	8.041	19.260
Transport by train	Electricity from national grid	0.879	2.106
Transport by 8t-truck	Gasoline	17.900	42.874
Processing
	Gasoline	5.925	14.192
	Natural gas	8.469	20.284
Total		41.750	100

. The carbon footprint of OC-derived products varied markedly across the three supply chains, reflecting inherent differences in system boundaries, functional units, input intensities, and emission-driving processes. Based on cradle-to-grave inventory modeling, the product-specific global warming potentials (GWPs) were quantified as 244.314 kg CO₂e/m³ for fibreboard (Table 4), 481.626 kg CO₂e/t for air-dried pulp (Table 5), and 41.750 g CO₂e/kWh for biomass-based electricity (Table 6), respectively. These results revealed that while the pulp pathway showed the highest absolute carbon intensity per unit of product, the fibreboard route concentrated the most emissions per unit of biomass utilized, and the bioelectricity system—benefiting from fuel substitution and limited material complexity—achieved the lowest unit carbon intensity (Table 3).

A closer examination of upstream plantation management highlighted a consistent baseline contribution of 11.821 kg CO₂e/m³ for wood fiber supply, primarily from diesel combustion, internal transport, and minor non-CO₂ emissions (e.g., CH₄, NO_x, BOD). This stage remained relatively uniform across all three products but played a proportionally larger role in the bioelectricity system, where processing stages involved simpler infrastructure and fewer chemical transformations (Table 3). For fibreboard production, energy use dominated the footprint (59.3%), particularly diesel (74.316 kg CO₂e/m³) and grid electricity (34.077 kg CO₂e/m³) (Table 4). In contrast, pulp manufacturing was heavily influenced by chemical consumption (231.480 kg CO₂e/t, 48.06%), especially CaO, NaOH, and H₂O₂, alongside site-generated electricity and fuel combustion (e.g., coal, bunker oil) (Table 5). These differences underscore the higher process complexity and input diversity of chemical pulping relative to thermo-mechanical board production. The bioelectricity system presented a distinct emission pattern. Gasoline use during forest residue collection, transport, and combustion accounted for nearly 60% of the total emissions, followed by natural gas co-firing (20.3%) and labor energy (19.3%). Despite the relatively modest emission intensity (41.750 g CO₂e/kWh), transportation and processing stages remain non-negligible, especially under decentralized supply networks (Table 6).

Collectively, these findings emphasize the importance of supply chain disaggregation and emission attribution in carbon accounting. Emissions were not solely determined by final energy or material output but were strongly modulated by the combination of activity types (e.g., combustion, chemical conversion, electricity use), material throughput, and energy sources. Moreover, the magnitude and distribution of emissions across upstream and downstream stages varied significantly between product systems. This highlights the need for tailored mitigation strategies: optimizing chemical recovery in pulping, reducing diesel dependency in board manufacturing, and enhancing transport efficiency in bioelectricity logistics. Ultimately, a product’s carbon profile reflects not just what it is, but how—and through what—it is made.

3.3. Emission Source and Activity Footprint: Fibreboard

Figure 2 provides a comprehensive overview of the carbon emission profile of OC fibreboard production, with a particular focus on the contributions from different stages of the production process (Tables S5–S8). As seen across Figure 2a–d, emissions are distributed across plantation management, energy use, material inputs, and processing activities, each with distinct contributions to the overall environmental impact.

The carbon emission profile of OC fibreboard production reveals significant variation across different stages of production. As shown in Figure 2a, the total global warming potential (GWP) is heavily concentrated in mill-stage activities, with 92.9% of emissions (227.007 kg CO₂e/m³) originating from the fibreboard mill. In contrast, plantation management contributes only 7.1% of the total emissions. This stark imbalance reflects the higher energy and material demand associated with engineered wood processing compared to upstream biomass production, where simpler and less energy-intensive operations are involved.

Zooming in on the composition of emissions (Figure 2b, Table S6), energy and fossil fuels stand out as the dominant contributors, accounting for 59.2% of the total GWP. This is followed by other materials (primarily water use) at 33.3%, with wood fiber raw materials contributing just 7.1%. While emissions and chemical use during processing are minimal (<0.5%) in terms of their direct carbon impact, they could be environmentally significant in other categories such as toxicity or eutrophication, which warrants further investigation.

The energy-related footprint (Figure 2c, Table S7) is particularly diverse, with diesel usage accounting for more than half (51.3%) of the energy-related greenhouse gas emissions. Diesel’s role in mechanized refining and thermal treatment of materials contributes significantly to these high emissions. Grid electricity comes in second at 23.5%, highlighting the embedded carbon intensity of China’s national electricity grid. Other energy sources, including liquefied petroleum gas (8.4%), natural gas (6.8%), and on-site generated electricity (5.6%), collectively reflect a hybridized energy portfolio, which contributes to the operational flexibility of fibreboard production facilities.

When examining the contribution from wood fiber materials (Figure 2d, Table S8), harvesting activities emerge as the primary emission driver, accounting for 69.8% of emissions. This stage involves high diesel consumption for felling and forwarding, further compounded by 21.8% emissions from internal transportation and 8.4% from direct emissions such as N₂O and CH₄ released during the disturbance of residues or exposure of soils. These results highlight the spatial and mechanical footprint of biomass provisioning in a plantation-based model, which, although designed to be sustainable, still exhibits significant environmental impact due to intensive machinery use.

Taken together, these findings present the fibreboard production system as one that is driven primarily by energy usage and the complexity of mill-stage processes. Despite the relatively minor contribution from upstream activities, such as plantation management, the emissions are still significant. The breakdown of mill emissions (Figure 2c) and upstream sourcing (Figure 2d) provides valuable insights for the design of targeted decarbonization strategies, such as reducing reliance on diesel and improving the electricity grid mix for more sustainable fibreboard production.

3.4. Emission Source and Activity Footprint: Pulp

The carbon emissions from pulp production are highly influenced by a combination of chemical use, energy consumption, and direct emissions from various stages of the process (Figure 3, Tables S9–S12). Dissecting the emission sources further (Figure 3b), chemicals emerge as the largest contributor, responsible for 48.1% (231.48 kg CO₂e/t) of the total emissions. These are followed by energy and fossil fuels (19.6%, 94.221 kg CO₂e/t), emissions to water (11%) and air (7.1%), feedstock materials (6.7%), smaller fractions from solid waste (6.4%) and water use (1.2%). The carbon footprint structure here contrasts sharply with fibreboard systems, where energy carriers play the dominant role, highlighting pulp’s inherently chemical-intensive character (Table 5).

A closer look at fossil energy contributions (Figure 3c, Table S11) reveals a diversified portfolio: on-site electricity generation—primarily through cogeneration—accounts for 31.2% of fossil energy-related emissions, followed by bunker oil (25.6%), diesel (18.8%), and coal (16.7%). Despite the presence of national grid electricity (6.3%) and internally used steam (1.4%), the system leans heavily on direct fuel combustion. This reliance on multiple energy vectors likely stems from the thermal and mechanical demands of cooking, bleaching, and drying stages, indicating opportunities for fuel substitution or efficiency upgrades.

The chemical profile (Figure 3d, Table S12) further reinforces the system’s process intensity. Calcium oxide (CaO), used extensively in digestion and pH regulation, alone contributes 43.9% of chemical emissions. Sodium hydroxide (NaOH), hydrogen peroxide (H₂O₂), and sodium chlorate (NaClO₃)—central to pulping and bleaching—collectively account for another 44.0%. Although recovered NaOH (8.9%) helps mitigate fresh input demand, its effect on overall emissions remains limited. Minor inputs like methanol (2.5%) and sulfuric acid (0.7%) contribute relatively little but signal the breadth of auxiliary chemical use in maintaining oxidation balance and post-treatment.

Taken together, these results point to a process with high conversion efficiency but at the cost of substantial embedded emissions from reagents and energy. The dominance of the pulp mill stage, particularly its chemical footprint, suggests that mitigation efforts should prioritize cleaner chemical alternatives, waste chemical recovery, and renewable heat sources. This integrated understanding of source-specific emissions lays the groundwork for evaluating substitution strategies and eco-efficiency improvements across the pulp supply chain (Table 5).

3.5. Emission Source and Activity Footprint: Bioelectricity

The carbon emission profile of OC-based bioelectricity follows a markedly different logic from that of material-oriented pathways, reflecting its terminal-use, combustion-driven nature (Figure 4, Tables S13–S15). The carbon emission profile of OC-based bioelectricity follows a markedly different logic from that of material-oriented pathways, reflecting its terminal-use, combustion-driven nature. As shown in Figure 4a (Table S13), the majority of GHG emissions (79.5%) are associated with the bioelectricity generation stage, while upstream plantation management accounts for only 20.5%. This sharp imbalance underscores the relatively simple structure of raw biomass collection and the centralized intensity of energy conversion facilities (Table 6).

A closer breakdown of emission contributions by life cycle activities (Figure 4b, Table S14) further reveals the logistics-centric nature of the system. Transportation from forest sites to energy facilities constitutes the single largest contributor, accounting for 45.0% of total GWP. This is closely followed by processing (34.5%), which includes combustion stabilization, feedstock handling, and power generation, while field-level operations such as harvesting and internal transport contribute a comparatively modest 20.5%. This partitioning indicates that although the feedstock is of low commercial value, its distributed spatial availability imposes significant carbon burdens through fuel and infrastructure use (Table 6).

The decomposition of input-based emission sources (Figure 4c, Table S15) paints a more granular picture of these operational drivers. Gasoline use overwhelmingly dominates the footprint (59.2%), primarily consumed during long-distance transportation and field equipment operations. Natural gas accounts for 20.3% and is used for ignition and temperature stabilization in small-scale boilers. Labor inputs, despite being a human factor, contribute a notable 19.3%, reflecting the manual-intensive nature of decentralized biomass mobilization. Electricity drawn from the national grid plays a minor role (1.3%), aligning with the relatively self-sufficient and low-electrification characteristics of rural biomass conversion systems.

Together, these patterns delineate a system shaped by logistical dispersion and minimal preprocessing. The carbon footprint of bioelectricity is not rooted in material transformation or chemical conversion, but rather in the fossil energy required to mobilize, deliver, and thermally stabilize low-density, spatially scattered biomass. This contrasts with the mechanized, chemically intensive pathways of fibreboard and pulp, positioning bioelectricity as a more spatially constrained yet process-simplified alternative in the OC resource cascade.

3.6. Sensitivity and Scenario Analysis

To assess the decarbonization potential of targeted interventions, three scenario-based modifications were applied across fibreboard, pulp, and bioelectricity systems (Figure 5, Tables S16–S21). While all pathways exhibited measurable reductions in GHG emissions, the scale and drivers of mitigation varied significantly across systems. In the fibreboard system (Figure 5a), where upstream logistics and grid electricity dominate emissions, replacing regional electricity with certified green power yielded a moderate GWP reduction of 12.8%. In contrast, shifting to localized biomass sourcing and electrified transport (Figure 5d) achieved a more substantial 37.2% reduction, highlighting the high carbon intensity of diesel-based collection and transport. This upstream sensitivity contrasts sharply with the pulp production pathway (Figure 5b), where emissions are predominantly process-driven. Here, substituting quicklime with lower-carbon hydrated lime resulted in a 6.9% decrease in GWP, while closed-loop wastewater treatment (Figure 5e) produced a more pronounced reduction of 17.9%, indicating the disproportionate impact of effluent-related emissions in chemical pulping systems.

The bioelectricity system (Figure 5c), with a lower baseline emission intensity, showed more incremental yet relevant improvements. Replacing fossil-derived natural gas with biogenic methane achieved an 18.4% reduction, and labor-related emission cuts through automation or restructuring contributed an additional 12.5% (Figure 5f). While the absolute GHG reduction was smaller than in material-intensive systems, the sensitivity of bioelectricity to fuel structure and non-material components underscores its systemic flexibility in low-tech contexts.

Taken together, the scenario analysis reveals divergent mitigation leverage points: transport and logistics in fibreboard, material and effluent management in pulp, and energy and labor restructuring in bioelectricity. These contrasts emphasize the need for pathway-specific interventions, rather than one-size-fits-all strategies, when designing carbon-efficient transitions in biomass-derived product systems.

4. Discussion

4.1. Carbon Emissions

Compared to published fibreboard studies (Table S22), the carbon footprint of OC fibreboard was significantly lower than that of Brazilian MDF (759 kg CO₂e/m³) [16] and US particleboard (720 kg CO₂e/m³), both of which were cradle-to-grave assessments that included high resin loads and fossil energy reliance [59]. This difference is largely attributed to the lower resin content in the OC system, its exclusion of urea-formaldehyde adhesives, and a greater reliance on grid electricity rather than on-site fossil combustion. The value also remained markedly below that of Chinese plywood (357–538 kg CO₂e/m³) under comparable boundaries, further highlighting the role of product form and energy mix structure [15,60]. For pulp, the gate-to-grave GWP of 481.626 kgCO₂e/t for OC chemi-mechanical pulp approximates the global averages for kraft and CTMP pulp (508–513 kg CO₂e/t) [61], but it remains below values reported for SBSK kraft pulp (722 kg CO₂e/t) [62]. Allocation-sensitive variations in tissue pulp LCAs ranged from 388 to 448 kg CO₂e/t, depending on methodological assumptions [63]. Notably, the relatively high GHG contribution from chemicals such as CaO and NaOH in this study accounted for nearly half of the total emissions, reinforcing the sensitivity of pulp systems to chemical dosing intensity and co-product allocation. In addition, the inclusion of on-site power generation and steam reuse helped mitigate external energy dependence, partially offsetting upstream burdens. OC bioelectricity exhibited the lowest carbon footprint among the three products, at 41.750 g CO₂e/kWh, under a gate-to-gate boundary. This result is significantly lower than values for sawdust-based pellets (0.14–0.56 kg CO₂e/kWh) [64] and US wood residues (0.138–0.173 kg CO₂e/kWh) [65,66], and even below hybrid LCA estimates of biomass power in China (0.068–0.087 kg CO₂e/kWh) [67,68]. The low carbon intensity can be attributed to high-efficiency combustion, reduced pre-treatment steps, and minimized electricity input during fuel processing, as well as the low emission factor used for field-collected gasoline. However, gasoline consumption still dominated the footprint, indicating that transportation optimization remains a key mitigation opportunity.

Collectively, these comparisons underline the value of OC residues in low-carbon transitions. The fibreboard pathway demonstrated moderate GHG intensity but offered potential for carbon substitution through wood-based material uses. The pulp pathway revealed high chemical dependency and therefore potential for green reagent substitution and recovery enhancement. The bioelectricity pathway, while showing the smallest per-unit GWP, exhibited notable sensitivity to transportation fuel and infrastructure efficiency. Across all pathways, the distinct system boundaries and product characteristics underscore the need for pathway-specific decarbonization strategies and the avoidance of cross-pathway misinterpretations.

4.2. Carbon Footprint Activity

The analysis of carbon footprints across various production systems of OC-derived products, including fibreboard, pulp, and bioelectricity, reveals significant differences influenced by the type of energy used, raw material inputs, and the complexity of processing activities. A detailed comparison across multiple studies highlights both the factors contributing to higher carbon footprints in certain pathways and those that lead to lower emissions, offering insights into possible mitigation strategies (Table S23).

During the production process of fiberboard, the carbon emission of OC tree fiberboard is 244.314 kg CO₂/kg. The majority of the emissions (92.92%) come from the production stage, followed by the planting and management stage (7.08%). This is consistent with the research results of Zhang et al. (2017) [62]. This is consistent with findings from Hussain et al. (2017), who observed that energy use in the mill is the primary contributor, accounting for a significant portion of the carbon footprint [69]. In comparison, Bergman (2014) reported a higher carbon footprint for North American cellulosic fibreboard, with emissions reaching 477 kg CO₂/m³ [70]. This higher value can be attributed to a more energy-intensive manufacturing process, which involves not only the direct processing of wood but also the higher electricity demand from the national grid, which is typically carbon-intensive in certain regions. These differences highlight the critical role of energy sources and production efficiency in shaping the carbon intensity of fibreboard production.

The carbon footprint of OC-derived pulp (481.626 kg CO₂e/t) is in line with conventional pulp production systems, where chemical use and fossil fuel consumption dominate emissions. Sahoo et al. (2021) report similar findings, identifying chemicals like CaO, NaOH, and H₂O₂ as key emission drivers in pulp production [71]. In comparison to traditional systems, such as the UFS pulp process with emissions of 1734 kg CO₂e/t [72] and CFS systems reaching 2064 kg CO₂e/t [73], the OC pulp shows lower emissions, primarily due to its reduced energy demands and more efficient chemical use. Furthermore, the OC system benefits from internal steam generation and energy recovery, minimizing its dependence on fossil fuels like coal and bunker oil, commonly used in pulp production, which contributes to high carbon intensity in systems like those studied by Piekarski et al. (2017) and McManus et al. (2015) [14,16]. Despite the chemical intensity, this study highlights the OC pulp production’s relative efficiency in emissions when compared to traditional methods. The lower emissions observed align with the growing emphasis on using more sustainable feedstock in pulp manufacturing, offering a promising alternative to conventional raw materials such as wood or wheat [74,75].

In the context of bioelectricity production from OC seed residues, Kadiyala et al. (2016) reported a relatively low carbon intensity [76]. The majority of emissions (79.46%) occurred during the bioelectricity generation phase, primarily due to biomass combustion. Whitaker et al. (2011) further highlighted that transportation was a major emission source, accounting for 45% of total greenhouse gas emissions—significantly higher than the transport-related contributions observed in fibreboard and pulp production systems [77]. This is particularly critical in biomass supply regions requiring long-distance transportation, a characteristic often inherent to bioelectricity supply chains [78]. In contrast, fibreboard and pulp systems generally involve more centralized production facilities and shorter transport distances, resulting in lower transport-related emissions. The disparity in transport emissions underscores the contrast between the logistical decentralization of bioenergy systems and the more centralized manufacturing processes of fibreboard and pulp.

The variation in carbon footprints across these product systems is largely attributable to differences in energy sources, raw material processing, and transportation logistics. For example, Buitrago-Tello et al. (2025) notes that energy use in pulp production is dominated by coal and bunker oil, which are both highly carbon-intensive [73]. Meanwhile, Sahoo et al. (2021) highlights that energy-related emissions from electricity and steam generation in pulp mills make up a significant portion of the carbon footprint, reflecting the heavy reliance on fossil fuels in many pulp production systems [71]. In comparison, bioelectricity production, while energy-intensive due to combustion processes, benefits from lower emissions during the processing stages, especially when compared to chemical-heavy pulp production.

In summary, differences in carbon footprints across these product pathways can largely be explained by energy consumption patterns, the chemical intensity of the processes, and transportation requirements. The use of fossil fuels, particularly coal and diesel, significantly contributes to the carbon intensity of pulp production, while fibreboard production is more dependent on energy-efficient processes, despite a heavy reliance on grid electricity in some regions. Bioelectricity, on the other hand, is heavily impacted by transportation emissions, which is a major source of variability in its carbon footprint. These findings suggest that a targeted approach, focusing on improving energy efficiency, transitioning to renewable energy sources, and optimizing logistics, can significantly reduce the carbon footprint of OC-based products.

4.3. Substitution Potential, Uncertainty and Research Implications

In assessing the substitution potential of OC residues for various products, it becomes evident that significant carbon reductions can be achieved in fibreboard, pulp, and bioenergy production. These reductions, however, vary widely depending on the type of product being replaced and the underlying process differences. By evaluating the carbon footprint reduction quantities from the provided data, we gain a clearer picture of where OC can have the most impact (Table S24).

Starting with fibreboard production, the use of OC as a raw material results in an emission of 244.314 kg CO₂e/m³, a considerably lower footprint compared to alternatives like softwood-based plywood, with emissions reaching 538 kg CO₂e/m³ [60]. This 82.7% reduction is primarily due to the less energy-intensive nature of OC processing, which contrasts with the higher energy demand for harvesting, transport, and wood treatments required by softwood and other materials [15]. In this case, the emission reduction from OC is a direct result of its simpler processing techniques and the lower need for external chemical inputs compared to conventional materials. This disparity in emissions is also seen when comparing OC to bamboo-based materials, which have even higher carbon footprints, especially in products like bamboo laminated floors (631.85 kg CO₂e/m³) [79]. Bamboo products require considerable energy for drying and treatment, contributing to their elevated carbon footprints, highlighting the significant potential of OC residues for reducing environmental impact in the panel industry.

Turning to pulp production, the substitution of traditional pulp materials with OC also yields impressive results. The emission factor for OC pulp is 481.626 kg CO₂e/t, markedly lower than that of eucalyptus pulpwood, which emits 1245.576 kg CO₂e/t [36]. This stark contrast can be attributed to the more efficient biomass conversion in OC compared to eucalyptus, where the latter’s carbon footprint is driven up by the energy-intensive processes associated with harvesting and pulping, as well as the significant use of chemicals [80]. Additionally, the first system of imported waste paper used in pulp production emits 826.54 kg CO₂e/t, highlighting the impact of raw material sourcing on overall emissions [81]. The high emissions in recycled paper production are largely due to the energy demands of paper reprocessing, particularly in the case of materials with higher impurity levels [81]. Conversely, OC offers a more sustainable alternative with a lower carbon intensity and less reliance on chemical additives, making it an attractive option for environmentally conscious pulp production.

In the realm of bioenergy, OC biodiesel stands out with an emission factor of 41.750 g CO₂e/kWh, offering a reduction potential when compared to other biofuels. For example, biomass power and thermal power, though renewable, exhibit higher emission factors of 57.4 g CO₂e/kWh and 1079.3 g CO₂e/kWh, respectively [82]. The higher emissions from biomass power stem from the use of agricultural and forestry residues, which involve transportation and pre-treatment processes that contribute additional carbon. However, OC biodiesel, benefiting from its more localized cultivation and reduced processing needs, can achieve a much lower carbon intensity. Wind power and solar power, both renewable sources, achieve even lower carbon footprints of 11.25 g CO₂e/kWh and −5.7 g CO₂e/kWh, respectively, owing to their clean energy production mechanisms and minimal material input requirements [82]. These findings underscore the importance of feedstock choice and energy system configuration in determining bioenergy’s carbon reduction potential.

The uncertainties in substitution potential arise from the variability in raw material sourcing, transportation logistics, and energy mix. For example, the OC biodiesel footprint is sensitive to cultivation practices, while other biofuels, like biomass and thermal power, are influenced by the carbon intensity of grid electricity and the type of feedstock used. The carbon footprint reduction for biomass power, although significant in some contexts, becomes less favorable when compared to other renewable energy sources due to the transportation and preprocessing involved, which adds substantial emissions [83,84]. Moreover, the use of wind and solar power, which have almost negligible emissions in comparison, highlights the discrepancy in the substitution potential when replacing fossil fuels with biofuels or bioenergy.

The implications of these findings are clear: substituting OC residues for traditional materials can lead to substantial carbon reductions, especially in industries like fibreboard and pulp production, where the emissions from conventional raw materials are considerably higher. The reduction in emissions is primarily due to the more efficient processing and less energy-intensive nature of OC, compared to alternatives like softwood, bamboo, and eucalyptus. However, achieving meaningful reductions in bioenergy emissions requires careful consideration of the energy mix and feedstock sourcing. Biomass power in particular faces challenges due to its reliance on agricultural residues, which necessitate transportation and pre-treatment processes that add to the carbon burden.

In conclusion, the substitution of OC residues for conventional materials offers significant emission reductions across fibreboard, pulp, and bioenergy sectors. However, the magnitude of these reductions varies depending on the feedstock and production methods. While the OC pathway presents a promising opportunity for carbon emission reduction, the overall impact is contingent upon the adoption of more efficient processing technologies and a shift toward cleaner energy sources. Further research into the scalability of OC-based production systems, as well as technological advancements in carbon capture and energy efficiency, will be essential for maximizing the environmental benefits of this approach.

4.4. Uncertainties and Implications

Although this study employed a process-based LCA framework supported by site-specific inventory data, several sources of uncertainty remain. These uncertainties primarily stem from regional variability in emission factors, allocation assumptions, and boundary definitions [85]. For instance, the electricity grid mix used in baseline scenarios may not accurately reflect local decarbonization progress in regions with large shares of hydropower or renewables, such as Yunnan or Sichuan [86]. This spatial inconsistency can influence the reliability of fibreboard and pulp carbon intensity estimations, as also noted in prior regional LCA studies in China’s forestry sector [87].

A second layer of uncertainty is introduced by the allocation method for OC residues. While this study applied energy-based allocation, other approaches such as mass- or economic-based methods may yield different attributions of environmental burdens among oil, pulp, and fibreboard products. Comparative research (e.g., Sahoo et al., 2021) suggests that allocation strategy can shift carbon footprint outcomes by more than 20% in multi-output systems, underscoring the importance of methodological transparency [71]. In the bioelectricity pathway, assumptions on combustion efficiency, bottom ash disposal, and labor-related emissions also introduce variability. Although labor emissions are marginal in magnitude, their inclusion reflects the granularity required under China’s updated carbon accounting guidelines [88], which call for full-chain traceability in forestry-based carbon credits.

From a policy perspective, these uncertainties offer practical insights rather than limitations. Under China’s dual-carbon goals—peak carbon by 2030 and neutrality by 2060—the Forestry and Grassland Administration and National Development and Reform Commission have emphasized the importance of integrating forestry residues into regional bioeconomy strategies [10,89,90]. The current study shows that carbon savings from OC residues are highly sensitive to transport distance, process energy structure, and end-of-life treatment. These insights align with policy calls for localized biomass supply chains, life cycle-oriented carbon reduction, and “carbon sink + bio-product” hybrid valuation mechanisms.

Compared with existing LCA studies on wood-based products [91,92], this study contributes a feedstock-specific perspective rooted in the underexplored OC sector. Unlike typical forestry LCA research that generalizes biomass pathways, the present analysis disaggregates three distinct product systems, quantifies process-level emission drivers, and models scenario-specific mitigation effects. This layered approach not only enhances attributional clarity but also provides policymakers with system-specific guidance, especially in regions where OC is co-managed for edible oil, timber, and energy purposes. Importantly, the sensitivity analysis reveals that no universal mitigation solution applies across all pathways. Fibreboard systems are primarily constrained by transport fuels, pulp systems by wastewater emissions, and bioelectricity by fuel sourcing. Therefore, policies should avoid one-size-fits-all incentives and instead adopt modular carbon mitigation frameworks that reward targeted process optimizations and life cycle integration. Such frameworks would enhance the environmental credibility of China’s forestry-based bioindustry while aligning with national MRV (monitoring, reporting, verification) standards under Article 6 of the Paris Agreement [93].

4.5. Future Needs

While this study confirms the low-carbon potential of Camellia oleifera-derived fibreboard, pulp, and bioelectricity, realizing systemic benefits requires more than process-level optimization. The heterogeneity of biomass—ranging from bark and shells to twigs and leaves—means that treating it as a homogeneous input risks inefficient use and inaccurate emission estimates. Similarly to how environmental impacts of materials depend on bioavailable fractions rather than bulk presence, as shown in CuO nanoparticle toxicity studies, future LCA models must better represent the functional diversity within biomass flows [94]. Moreover, cascading utilization of lignin, as demonstrated through pyrolysis into both monophenolic chemicals and adsorbent materials [95], offers a model for extracting higher value from residues beyond single-output pathways. Also, future studies could involve considering additional environmental impact categories beyond just carbon footprint, such as acidification potential (AP) eutrophication potential (EP), photochemical ozone creation potential (POCP), human toxicity potential (HTP), fossil depletion potential (FDP), etc. AP could be calculated by summing contributions of SO₂, NO_x, and NH₃ emissions weighted by their acidifying equivalence (e.g., kg SO₂-eq), while EP included emissions of NO₃⁻, NH₃, and COD weighted in terms of kg PO₄³⁻-eq. POCP captured the reactivity of VOCs and ozone precursors such as CO and NO_x during biomass combustion and resin application, while HTP reflected the cumulative toxic equivalence of heavy metals, formaldehyde, and persistent organics. FDP quantified non-renewable resource use based on embedded energy in fossil-based electricity, diesel, and chemical inputs. Integrating such strategies into camellia systems could simultaneously enhance carbon efficiency, product value density and environmental sustainability.

Efforts to scale camellia-based industries must also account for spatial planning, predictive precision, and financial mobilization. Emerging optical analysis tools, such as multispectral bidirectional reflectance techniques [96], can improve yield prediction and feedstock matching, reducing mismatch between supply and processing demand. Meanwhile, access to sustainable finance remains pivotal. Empirical evidence from China’s energy sector shows that green bond issuance significantly boosts green innovation by easing financing constraints and encouraging technology upgrades [97]. Embedding these insights into policy design could support broader diffusion of camellia-based low-carbon products, making them competitive alternatives in national climate and industrial strategies.

5. Conclusions

The life cycle assessment (LCA) of products derived from OC seed residues—namely fibreboard, pulp, and bioelectricity—reveals considerable potential for carbon emission reduction across various production pathways. The life cycle carbon footprints of OC-derived products exhibited substantial divergence, reflecting distinct system boundaries, material flows, and process characteristics. Specifically, the cradle-to-grave carbon intensity of fibreboard reached 244.314 kg CO₂e/m³, while that of air-dried pulp was 481.626 kg CO₂e/t, and that of bioelectricity was 41.750 g CO₂e/kWh. These results not only reflect product-specific functional units but also the composite influence of energy structure, material input intensity, and emission factor (EF) selections. While OC seed residues exhibit lower carbon intensity in both fibreboard and pulp production, especially when replacing traditional wood materials, pulp production remains carbon-intensive due to chemical use and fossil fuel consumption. Conversely, bioelectricity generation, although associated with lower emissions, is more sensitive to transportation logistics and fuel type, particularly in decentralized energy systems. Scenario and substitution analyses suggest that emission reductions can be significantly achieved through strategic optimizations, including improvements in energy mix, substitution of high-carbon chemicals, and enhancements in transportation efficiency. However, the emission reduction potential varies across the different production pathways, with each facing specific challenges. For instance, fibreboard production benefits from a reduced reliance on diesel and electricity, while greater emission reductions in pulp production could be achieved through chemical substitutions and improvements in wastewater treatment. These findings underscore the promising potential of integrating OC seed residues into existing production systems, offering a viable path toward a low-carbon transition in the wood product industry. To fully capitalize on this emission reduction potential, further optimization of production processes, better resource utilization, and the establishment of favorable policy frameworks are critical for large-scale implementation.

6. Patents

Zhao, M.; Yao, T.; Xiong, T.; Lu, F.; Kang, P.; Wei, L.; Xia, Y. Carbon footprint accounting method for products produced from the transformation residues of low-yield oil camellia forests. CN202410975444.0, filed 19 July 2024, published 15 November 2024. Available online: https://www.iprdb.com/patent/CN118967152A.html (accessed on 21 July 2025).