The Effect of Forest Growth Rate on Climate Change Impacts of Logging Residue Utilization

Abstract

Biofuel is encouraged because of its low impact on climate change. A new framework was developed to accurately assess the climate change impacts (CCI) of biofuel by integrating the atmospheric carbon cycle model and vegetation carbon dynamic models. Forests with different growth rates (fast, medium, slow) and three collection intensities (71%, 52%, 32%) of logging residues were presumed to test the performance of this framework. The CCI of biofuel was analyzed under two functional units: 1 GJ of biofuels and 1 ha of forests to supply biofuels. According to this study, increasing the forest growth rate could decrease the CCI in both functional units. Increasing the collection intensity could decrease the CCI of 1 GJ of biofuel but increase the CCI of 1 ha of forest land (unless the impacts were negative in fast-growth forests with high and medium collection intensities). Producing bioethanol resulted in a lower CCI (−3.1–67.7 kg CO₂ eq/GJ) compared to bio-diesel (29.3–94.7 kg CO₂ eq/GJ). Hence, collecting all available logging residues (without inhibiting forest regrowth) to produce low CCI biofuels such as bioethanol was found to be the optimal option for achieving high mitigation effects.

1. Introduction

Rapid climate change has caused irreversible damage to the atmosphere, oceans, cryosphere, and biosphere. Anthropogenic greenhouse gas emissions are an unequivocal reason leading to global warming [1]. The global average atmospheric CO₂ concentration has increased by 49% from 1750 to 2020, and in 2021, global CO₂ emissions from fossil fuels will reach 36.8 Gt, accounting for 89% of all anthropogenic CO₂ emissions [2]. To proactively address climate change, adapt to its impacts, and achieve sustainable development strategies, the Paris Agreement called for limiting global warming to below 2 °C, with a preference for limiting it to below 1.5 °C in this century [3]. Mitigating global warming is an urgent issue that needs intensive international collaboration [4].

To fulfill the commitment to reduce carbon emissions [3,5], the production and use of biofuels have increased continually in many countries from 2012 to 2022 [6] because of their lower life-cycle greenhouse gas emissions than conventional fuels [7,8]. Political encouragement is a strong guarantee for the advancement of biofuel technologies, such as blending mandates, excise tax exemptions and incentives, renewable or low-carbon fuel standards, fiscal incentives, and public financing [9]. Biofuel feedstock can be sourced from a variety of sources; forest logging residues are an important source of biomass energy [10]. Logging residues from forests can provide 2–4% of biomass demand [11], and demand for logging residues will grow rapidly in the coming decades as the demand for biomass energy increases [12]. It is expected that by 2031, woody residues will account for 18–26% of the biomass used for bioenergy [13]. Logging residue utilization does not create food competition problems [14], and the removal of logging residues reduces the probability of wildfires [15]. Moreover, long-life cycle lignocellulosic feedstock requires less fertilizer and pesticides and less human intervention, resulting in a lower impact on biodiversity than short-rotation plants, such as crops [16].

Plant-based biofuel combustion is often considered to have zero CO₂ emissions [17], as biomass-derived CO₂ emissions can be absorbed through regrowth and are assumed to be carbon neutral [18]. However, the short duration of biomass-derived CO₂ emissions and the time required for biomass regrowth to compensate for these emissions could have significant climate change impacts (CCI) that should be considered [19,20]. When considering the Global Warming Potential (GWP) of biomass-derived CO₂ emissions and the compensation effect, the CCI from biofuel could be significantly higher than that of fossil fuels [10]. Furthermore, the impact on land use changes should be treated as a direct consequence of biomass utilization [21,22]. Thus, the carbon-neutral assumption may not hold when the direct consequences of biomass usage are comprehensively considered and integrated.

Given the non-carbon neutral assumption of plant-based biofuel, researchers have sought to develop new methods to accurately assess the CCI of biofuel, maximize its mitigation benefits, and ensure the sustainable utilization of biomass by integrating, such as the stay of biomass-derived CO₂ emissions in the atmosphere, the effect on carbon sequestration [19,23,24]. The GWP is commonly used as a reference indicator in life cycle assessment (LCA) research [25]. Based on these considerations, Liu et al. [26] developed a new LCA framework for assessing the impacts of biomass use on climate change by integrating the atmospheric carbon cycle model and the dynamic vegetation model. This framework was modified by Hao et al. [27], and the updated framework included four components: fossil fuel-derived GHG emissions, biogenic GHG emissions, biomass regrowth for compensation, and the difference in carbon sequestration to the reference scenario. This framework enables a fair comparison to fossil fuels and promotes the sustainable utilization of biomass [26,27,28,29].

The forest growth rate has a direct impact on the potential to mitigate climate change when harvesting forest logging residues for biofuel [30]. The rate of regrowth and decomposition of logging residues varies among different forests [31], which can impact biomass regrowth for compensation and the difference in carbon sequestration compared to the reference scenario. Some previous studies have used this new LCA framework to estimate the CCI of biomass for bioenergy [27,32,33], but to date, no one has assessed the climate change impacts of biofuel from logging residues at different forest growth rates using the newly developed LCA framework.

Therefore, the objectives of this study are: (1) to integrate carbon dynamics models into this new framework to estimate the CCI of forest logging residue for bioethanol and bio-diesel; (2) to compare the differences in the CCI of biomass from forests with different growth rates.

2. Methods and Materials

2.1. Scenario Definition

In this study, 18 scenarios were defined based on three forest growth rates, three collection intensities, and two types of biofuel. Three forest types were chosen to represent three different growth rates: boreal coniferous forests represent slow-growing forests, temperate continental forests represent medium-growing forests, and tropical rainforests represent fast-growing forests. After forest harvest, logging residues are typically left in the forests and decompose slowly over time. The proportion of available logging residues (as a percentage of aboveground biomass) in different forests was 28% for boreal coniferous forests, 35% for temperate continental forests, and 41% for tropical rainforests [18]. The collected logging residues were presumed to be converted into two types of biofuels: bioethanol and bio-diesel.

The fractions of collectible logging residues for biofuel were set proportionally to avoid inhibiting forest regrowth based on previously published studies: high intensity at 71%, medium intensity at 52%, and low intensity at 32% [34,35]. Different collection intensities are set to assess the climate mitigation effect of logging residues from different forest growth rates to obtain the optimal utilization strategy. Furthermore, the rich scenario settings are more conducive to testing the performance of the new LCA framework. Eng [34] mentions that the actual collection intensity can only be 60% but assumes that the technology and equipment are upgraded to set the ideal maximum collection intensity at 70%. Therefore, in this study, we set the maximum threshold of collection intensity at 71%, even though a higher level may yield a better climate change mitigation effect.

2.2. Assessment of Climate Change Impacts

Life cycle assessment (LCA) is a standard method for evaluating the environmental impacts of a product and has been widely used in the CCI assessment of biofuel. In this study, a cradle-to-grave LCA model was developed, encompassing logging residue collection, transport, biofuel production, distribution, and final combustion stages. Two functional units were employed in this study: 1 GJ of energy equivalent biofuel to enable comparisons among different energy products and 1 ha of forestland for biomass collection to provide insights for forest management and land use strategies. We use the allocation method in the LCA of logging resources used for bioenergy [36].

Climate change impacts were assessed using the framework developed by Hao et al. [27]. The assessment time horizon (T) was 100 years, and all climate change impacts were calculated in CO₂ equivalents. The assessment framework consists of four components (Figure 1): fossil fuel-derived GHG emissions ($E_{fossil}$, kg CO₂ eq), biogenic GHG emissions ($E_{bio}$, kg CO₂ eq), biomass regrowth for compensation ($E_{com}$, kg CO₂ eq), and the difference in carbon sequestration ($E_{diff}$, kg CO₂ eq). The total CCI of biofuel ($E_T$, kg CO₂ eq) is the sum of the four components (Equations (1) and (S1)).

$$E_T=E_{fossil}+E_{bio}+E_{com}+E_{diff}$$

(1)

A brief description of the calculation process for each component can be found in the following paragraphs. The functional unit of 1 GJ energy equivalent biofuel was used to illustrate the calculation process, and the CCI for the functional unit of 1 ha of forestland can be obtained by multiplying the amount of biofuel that can be produced from the forestland. To reduce the difficulty of understanding the framework and allow the readers to focus on the structure of the framework, detailed calculation formulas were provided in the Supplementary Information. And we have listed all abbreviations and its expressions in Table S2.

Of the four components, fossil fuel-derived GHG emissions ($E_{fossil}$, kg CO₂ eq) are typically well-accounted for in traditional LCA studies, and publicly available databases can be used [20]. In this study, fossil fuel-derived GHG emissions from all processes were obtained from the GREET 2021 database [37] (See Section 3.1 for specific values).

The biogenic GHG emissions ($E_{bio}$, kg CO₂ eq) represent the climate change impacts of biomass-derived CO₂ emissions. The biogenic GHG emissions are calculated as the product of the ${GWP}_{bio}$ and the initial biomass-derived CO₂ emissions within 100 years (E(0), kg C/ha) (Equation (2)). $\delta$ ha represents the land area needed to produce 1 GJ of energy equivalent from biofuel (Equation (S3)). ${GWP}_{bio}$ refers to the global warming potential of biomass-derived CO₂ emissions (Equation (S4)).

$$E_{bio}=\frac{44}{12}\delta E\left(0\right){GWP}_{bio}$$

(2)

The biomass regrowth for compensation ($E_{com}$, kg CO₂ eq) is the CCI of biomass-derived CO₂ emissions absorbed by forest regrowth, which is calculated as the sum product (COM) of $GWP\left(t\right)$ and $COM\left(t\right)$ within the time horizon (T = 100 years, Equation (3)). $GWP\left(t\right)$ is the global warming potential of CO₂ emitted in the atmosphere in year $t$ (Equation (S12)), and $COM\left(t\right)$ is the compensation in year $t$ (Equation (S13)).

$$E_{com}=\frac{44}{12}\delta {\int }_1^{T}GWP\left(t\right)COM\left(t\right)dt$$

(3)

The difference in carbon sequestration ($E_{diff}$, kg CO₂ eq) is the CCI of the different carbon sequestration between the logging residue utilization scenario and the reference scenario. The reference scenario was defined as retaining all the logging residues in the forests. In this study, if logging residues are collected for biofuel without impacting forest regrowth, the carbon in the residues is emitted as a one-time pulse from combustion, and logging residues naturally decompose if not harvested for biofuel. Therefore, $E_{diff}$ is calculated as the sum product ($C_C$) of $GWP\left(t\right)$ and the difference in carbon sequestration between the logging residue utilization scenario and the reference scenario in year t, within the time horizon of 100 years (Equations (4) and (S15)).

$$E_{diff}=\frac{44}{12}\delta C_C=\frac{44}{12}\delta {\int }_1^T(-1)·GWP\left(t\right)·D^{\prime }\left(t\right)$$

(4)

2.3. Simulation of Carbon Dynamics

Simulating the carbon dynamics of different forests is necessary to estimate the CCI of logging residues for biofuel production using the LCA framework. In this study, forest regrowth was simulated using the Chapman-Richards function (Section 2.3.1). The YASSO15 model was used to model the decomposition of uncollected logging residues (Section 2.3.2).

2.3.1. Forest Growth Simulation

The Chapman-Richards function was first published by Richards [38], based on the theory proposed by Von Bertalanffy [39], and later introduced to forestry by Pienaar and Turnbull [40]. Due to its flexibility and accuracy, the Chapman-Richards function is a commonly used model to describe the growth of various trees and stands [40,41]. For this study, the Chapman-Richards function was utilized to simulate forest growth with different growth rates (Equation (5)):

$$A\left(a\right)=b_1{(1-e^{-b_{2}a})}^{b_3}$$

(5)

where $A\left(a\right)$ represents the biomass accumulation of the stand in the year a, in kg C/ha, and $b_1$, $b_2$ and $b_3$ represents empirical parameters from previous studies. Simulation of forests with slow-growing rates in boreal coniferous forests, medium-growing rates in temperate continental forests, and fast-growing rates in tropical rainforests was conducted using three sets of parameters (

Table 1. Chapman-Richards function parameter settings.

Forest Types	Forest Growth Rates	${\mathit{b}}_1$	${\mathit{b}}_2$	${\mathit{b}}_3$	Citation
Boreal coniferous forests	Slow	103.1	0.0245	2.69	Holtsmark [42]
Temperate continental forests	Medium	198.6	0.0253	2.64	Holtsmark [42]
Tropical rainforests	Fast	428.0	0.0253	2.64	Asante et al. [43]

). Based on previous studies, the commercial harvest was scheduled when the accumulated biomass exceeded 80,000 kg C/ha [35]. Simulation results for forests with three different growth rates are presented in Figure S1a,b.

2.3.2. Decomposition of Logging Residues

To calculate the difference in carbon sequestration in the LCA framework, the decomposition of unused logging residues in all scenarios was simulated by the soil carbon model YASSO15. YASSO15 is a widely used and well-validated model for simulating the decomposition of dead organic matter [44,45]. The YASSO15 model divides logging residues into five chemical components: compounds that are hydrolyzable in acid (A), compounds that are soluble in water (W) or a non-polar solvent, such as ethanol or dichloromethane (E), compounds that are neither soluble nor hydrolyzable (N), and humus (H). In this study, the proportions of the five components of logging residues from the three forests were averaged based on the typical tree species in each forest type [46] (

Table 2. The initial input for C pools from five chemical components in YASSO for the three different forest types. A: hydrolyzable in acid; W: compounds that are soluble in water or a non-polar solvent; E such as ethanol or dichloromethane; N: compounds that are neither soluble nor hydrolysable; H: and humus (%).

Forest Types	Five Chemical Components (%)
	A	W	E	N	H
Boreal coniferous forests	0.668	0.018	0.006	0.308	0.000
Temperate continental forests	0.466	0.021	0.080	0.433	0.000
Tropical rainforests	0.466	0.021	0.080	0.433	0.000

).

For simulating the decomposition of logging residues using YASSO, three sites were randomly selected from the Global Ecological Zones map (http://www.fao.org/geonetwork/, accessed on 15 November 2022) for the boreal coniferous forest, temperate continental forest, and tropical rainforest, respectively. These nine sites were located across Asia, Europe, South America, North America, and Africa (Figure 2a).

Climate data, including average annual precipitation, −annual average temperature, and annual mean temperature difference, were obtained from the CRU TS4.05 climate database (https://crudata.uea.ac.uk/cru/data/hrg/cru_ts_4.05/, accessed on 18 December 2022) for use in YASSO15. Decomposition curves for logging residues are presented in Figure 2b–d.

3. Results

3.1. Fossil Fuel-Derived GHG Emissions

The fossil fuel-derived GHG emissions were retrieved from the GREET database. To produce 1 GJ of biofuel from logging residues, the fossil fuel-derived GHG emissions were 19.67 kg CO₂ eq/GJ for producing bioethanol and 50.53 kg CO₂ eq/GJ for producing bio-diesel [37]. When the functional unit was 1 ha, the fossil fuel-derived GHG emissions were derived from the biomass yield under different scenarios (fast-growing: 7257.2–16,102.0 kg C/ha; medium-growing: 9272.8–20,352.1 kg C/ha; slow-growing: 10,563.4–23,437.5 kg C/ha) and energy conversion efficiency, represented by $\rho$ (bioethanol: 0.0174 GJ/kg C logging residues; bio-diesel: 0.019 GJ/kg C logging residues). For the production of bioethanol, fossil fuel-derived GHG emissions ranged from 2492.4 to 8049.1 kg CO₂ eq/ha, while for the production of bio-diesel, the range was 6960.4 to 22,478.8 kg CO₂ eq/ha.

3.2. Biogenic CO₂ Emissions and Biomass Regrowth for Compensation

Biogenic CO₂ emissions were based on GWP_bio. The value of GWP_bio was affected by the collection intensity and growth rate. GWP_bio ranged from 0.12 to 0.27, which is significantly lower than the GWP of fossil fuels. High collection intensity and slow-growth rate led to high GWP_bio (

Table 3. The initial CO₂ emissions of biofuel, ${GWP}_{bio}$ (global warming potential of biomass-derived CO₂ emissions), the climate change impact (CCI) of the component of the biogenic CO₂ emissions, and biomass regrowth for compensation in different scenarios (Percentage of collection intensity: High–71%; Mid–52%; Low–32%).

Forest Growth Rate	Collection Intensity	Biomass-Derived CO2 Emissions (kg CO2 eq/ha)	GWPbio	Biogenic CO2 Emission (kg CO2 eq)				Regrowth for Compensation (kg CO2 eq)
				Bioethanol		Bio-Diesel		Bioethanol		Bio-Diesel
				1 GJ	1 ha	1 GJ	1 ha	1 GJ	1ha	1 GJ	1 ha
Slow-growing	High (71%)	23,437.5	0.27	56.6	23,163.4	52.1	23,163.4	94.2	38,551.3	86.7	38,551.3
	Mid (52%)	17,165.5	0.24	51.0	15,283.9	46.9	15,283.9	99.9	29,955.1	91.9	29,955.1
	Low (32%)	10,563.4	0.21	43.7	8063.1	40.2	8063.1	108.2	19,953.6	99.5	19,953.6
Medium-growing	High (71%)	20,352.1	0.20	41.9	14,900.8	38.6	14,900.8	110.4	39,244.5	101.6	39,244.5
	Mid (52%)	14,905.7	0.18	38.1	9916.4	35.1	9916.4	115.4	30,042.6	106.2	30,042.6
	Low (32%)	9172.8	0.16	33.0	5287.5	30.4	5287.5	122.9	19,675.4	113.0	19,675.4
Fast-growing	High (71%)	16,102.0	0.15	31.1	8752.9	28.6	8752.9	125.9	35,388.8	115.8	35,388.8
	Mid (52%)	11,793.0	0.14	28.5	5867.6	26.2	5867.6	130.4	26,853.7	120.0	26,853.7
	Low (32%)	7257.2	0.12	25.0	3171.1	23.0	3171.1	137.1	17,371.4	126.1	17,371.4

). The biogenic GHG emissions for the functional unit of 1 GJ ranged from 25.0 kg CO₂ eq/GJ to 56.6 kg CO₂ eq/GJ for bioethanol and from 23.0 kg CO₂ eq/GJ to 52.1 kg CO₂ eq/GJ for bio-diesel. Biogenic GHG emissions for logging residues from forest land of 1 ha were the same for both biofuel products (3171.1–23,163.4 kg CO₂ eq/ha).

The biomass-derived CO₂ emissions have a short stay in the atmosphere due to the compensation from forest regrowth; therefore, the compensation effect should be quantified and accounted for as CCI of biofuel. The biomass-derived CO₂ emissions can be fully compensated in a short period, and the length of this period was defined as the compensation period. Compensation periods for the slow-growing, medium-growing, and fast-growing forests were 19–27 years, 13–18 years, and 9–12 years, respectively (Figure 3). When the functional unit was 1 GJ, the biomass regrowth for compensation was 94.2–137.1 kg CO₂ eq/GJ for bioethanol and 86.7–126.1 for bio-diesel. For a functional unit of 1 ha, the biomass regrowth for compensation was 17,371.4–39,244.5 kg CO₂ eq/ha for bioethanol and bio-diesel (Table 3). In general, greater consideration should be given to biomass regrowth for compensation in faster-growing forests with lower collection intensities.

3.3. The Difference in Carbon Sequestration

The difference in carbon sequestration was determined by the annual difference in carbon dynamics (i.e., decomposition in this study) between the utilization scenarios and the reference scenario and discounted to present emissions. In this study, the difference in carbon sequestration was negative, which was critical in offsetting the positive CCI. For the production of bioethanol, the difference in carbon sequestration to produce 1 GJ ranged from −68.0 to −180.8 kg CO₂ eq/GJ, while for bio-diesel, the range was −62.5 to −166.3 kg CO₂ eq/GJ. Fast-growing forests have the most negative difference in carbon sequestration, followed by medium-growing forests and slow-growing forests. The difference in carbon sequestration to produce both biofuels was the same when the functional unit was 1 ha and ranged from 12,536.0 kg CO₂/ha to 53,353.0 kg CO₂/ha. This component had lower values in faster-growing forests with higher collecting intensity (

Table 4. The CCI of the component of the difference in carbon sequestration in different scenarios. (Percentage of collection intensity: High–71%; Mid–52%; Low–32%).

Forest Growth Rates	Collection Intensity	Bioethanol (kg CO2 eq)		Bio-Diesel (kg CO2 eq)
		1 GJ	1 ha	1 GJ	1 ha
Slow-growing	High (71%)	−103.9 ± 26.3	−42,505.3 ± 10,774.6	−95.5 ± 24.2	−42,505.3 ± 10,774.6
	Mid (52%)	−103.9 ± 26.3	−31,130.7 ± 7891.3	−95.5 ± 24.2	−31,130.7 ± 7891.3
	Low (32%)	−103.9 ± 26.3	−19,157.3 ± 4856.2	−95.5 ± 24.2	−19,157.3 ± 4856.2
Medium-growing	High (71%)	−124.3 ± 9.8	−44,154.1 ± 3465.8	−114.3 ± 9.0	−44,154.1 ± 3465.8
	Mid (52%)	−124.3 ± 9.8	−32,338.2 ± 2538.4	−114.3 ± 9.0	−32,338.2 ± 2538.4
	Low (32%)	−124.3 ± 9.8	−19,900.4 ± 1562.1	−114.3 ± 9.0	−19,900.4 ± 1562.1
Fast-growing	High (71%)	−180.0 ± 1.1	−50,616.4 ± 302.9	−165.6 ± 1.0	−50,616.4 ± 302.9
	Mid (52%)	−180.0 ± 1.1	−37,071.2 ± 221.8	−165.6 ± 1.0	−37,071.2 ± 221.8
	Low (32%)	−180.0 ± 1.1	−22,813.0 ± 136.5	−165.6 ± 1.0	−22,813.0 ± 136.5

).

3.4. Life-Cycle Climate Change Impacts and Mitigation Effect

The life-cycle CCI is the sum of the four components. For a functional unit of 1 GJ, the life-cycle CCI for bio-diesel (29.3–94.7 kg CO₂ eq/GJ) was larger than that for bioethanol (−3.1–67.7 kg CO₂ eq/GJ) when the growth rate of forests and collection intensity were the same (Figure 4). The slow-growing forest had the highest CCI, followed by medium-growing and fast-growing forests. Higher collection intensity resulted in a lower life-cycle CCI.

For a functional unit of 1 ha, the life-cycle CCI to produce bio-diesel (4689.9–41,688.1 kg CO₂/ha) was also larger than that for bioethanol (−944.9–27,258.5 kg CO₂/ha) when the other conditions were the same (Figure 4). When the collection intensity was the same, the highest CCI was obtained from the slow-growing forest, and the lowest was obtained from the fast-growing forest. For both bioethanol and bio-diesel, the highest life-cycle CCI was obtained with the lowest collection intensity, followed by medium and high collection intensities.

Bioethanol and bio-diesel were used to substitute light distillate fuels and middle and heavy distillate fuels, respectively. According to a range of GHG emissions from fossil fuels, we provide a range of climate mitigation effects from substituting fossil fuels with biofuels, which can reduce the uncertainties from demand for biofuels and the development of conventional energy efficiency in the future. The life-cycle CCI ranged from 95.2 to 117 kg CO₂ eq/GJ for energy-equivalent light distillate fuels and from 88.9 to 109.9 kg CO₂ eq/GJ for middle and heavy distillate fuels [11] (Table S1). I In this study, the life-cycle CCI of biofuel products was lower than the energy equivalent of fossil fuels, except for bio-diesel from slow-growing forests (Figure 4). With the production of every 1 GJ of biofuel, the mitigation effect was 27.5–120.4 kg CO₂ eq/GJ for bioethanol and −5.7–80.6 kg CO₂ eq/GJ for bio-diesel. When the functional unit was 1 ha, the production of bioethanol from logging residues reduced emissions by 5070.9–33,837.7 kg CO₂ eq/ha. The highest mitigation effect from bio-diesel was 18,202.4 kg CO₂ eq/ha in the fast-growing forest with high collection intensity.

4. Discussion

4.1. Model Justification

The CCI of biofuel from traditional LCA models was usually lower than substituted fossil fuels [11,39]; however, the traditional LCA models were biased by ignoring the renewable nature of biofuel and the consequences of biomass utilization. Biogenic CO₂ emissions are mainly CCI from biofuel due to their staying in the atmosphere [13,19]. Taking into account the CCI of the stranded biomass-derived CO₂ in LCA can provide more reliable results.

The CCI of the biomass regrowth for compensation should also be considered in the LCA because this component is subject to offset biomass-derived CO₂ emissions through carbon sequestration [20,27]. After considering this component, the CCI of the biomass-derived CO₂ emissions is usually lower if the biomass regrowth is accountable. The actual values of biogenic CO₂ emissions and biomass regrowth for compensation from logging residues for biofuel are directly affected by the forest growth rate and collection intensity of logging residues.

The differences in carbon sequestration should be taken into account in the LCA framework because the collection of logging residue avoided emissions from the decomposition of logging residues. The avoidance of emissions from decomposition can be a significant benefit of biomass utilization [30]. The value of this component is obtained by comparing the difference between biomass utilization and no utilization (reference scenario). This component is essential to ensuring the sustainability of logging residue removal by limiting the collection intensity within a range that can avoid decomposition and will not inhibit forest regrowth. The decomposition rate of the logging residues is closely related to the local climate [45]. Therefore, the difference in carbon sequestration is usually higher in wetter and warmer regions, which have higher decomposition rates [24]. Hao et al. [27] used this LCA framework to estimate and compare the climate change impact (CCI) of hybrid poplar use for bioethanol, bio-diesel, heating, and bio-power. The total CCI of different types of bioenergy ranged from −26.85 kg CO₂ eq/GJ to 1.08 kg CO₂ eq/GJ, and bioethanol has the lowest CCI, which is consistent with this article. Liu et al. [24] estimated the global CCI of forest residue for bioenergy and found a lower CCI than traditional LCA and enormous potential for carbon reduction (292.7 to 864.2 Tg CO₂ eq).

In this study, to compare the difference among the three forest growth rates, we put the spotlight on carbon dynamics in which only the forest growth rate is different. Therefore, we just used the Chapman-Richards function to demonstrate three types of forest growth (slow-growing, medium-growing, fast-growing) using three series of parameters, which is suitable for simulating boreal coniferous forests, temperate continental forests, and tropical rainforests. Although the results of such a simulation will deviate from the actual situation, it is a good intuitive response to the growth process of forests with different growth rates.

4.2. Climate Change Impacts and Mitigation Effects

The total CCI was obtained by summing up four components. The positive CCI came from fossil fuel-derived GHG emissions, biogenic CO₂ emissions, and biomass regrowth for compensation, while the negative CCI came from the difference in carbon sequestration. Guest et al. [47] also considered emissions from the natural decomposition of logging residues in the forest and simulated this part using the YASSO15 and set nine logging residue utilization percentages from 0% to 100% to obtain a GWP of 0.44–0.62 for biofuel. The range of GWP in this study was 0.12–0.27, which is significantly lower due to the consideration of land use change emissions, which are calculated by the difference in CO₂ emissions between the scenarios with and without using logging residues. Hammar et al. [30] found that logging residues located in the warm south would have a higher potential to mitigate temperature change but would produce greater greenhouse gas emissions if decomposed naturally in the forest compared to the northern and central regions, which is consistent with the results obtained in our study.

In our assessment, most of the scenarios had significantly lower GHG emissions than alternative fossil fuels, except for the scenario of producing bio-diesel with slow-growing forests. The result indicated that, although biofuels from logging residues in forests can have positive life-cycle climate change impacts, they will still have a good emission reduction effect by substituting fossil fuels. Nonetheless, if the use of logging residue can inhibit forest regrowth or cannot avoid decomposition, the life-cycle CCI of biofuel will significantly increase and be much higher than that of fossil fuels [26].

Two different functional units were used in this study. Using a functional unit of 1 GJ energy-equivalent biofuel products ensures an effective comparison of the CCI of biofuel and fossil fuels and an analysis of the mitigation effect of different biomass energization usage scenarios. When land area (1 ha) was used as a functional unit, the results from various biomass energization use scenarios could serve as a guideline for land use policy under the requirement of decarbonization. By comparing the two biofuels (i.e., bioethanol and bio-diesel) in different functional units, the CCI of 1 GJ bioethanol was smaller, which indicated a better mitigation effect on climate change than producing bio-diesel. The energy conversion efficiency $\rho$ for the two biofuels was not significantly different [37]; therefore, bioethanol would be the reasonable choice in this study to maximize the mitigation effect. However, the advance in high-efficiency and clean conversion technology is essential to both the mitigation of climate change and ensuring energy security.

Furthermore, the life-cycle CCI and mitigation effects from biofuels were influenced by both forest growth rate and collection intensity of logging residues because all four components were derived from carbon dynamics except the fossil fuel-derived GHG emissions. The effects of forest growth rate and collection intensity had similar performance on biogenic CO₂ emissions and biomass regrowth for compensation. For the difference in carbon sequestration, the values depended on the decomposition rate of unused logging residues in different forests. Therefore, when producing 1 GJ of biofuel, the difference in carbon sequestration was identical to producing the same biofuel product because the demand for feedstock had not changed. For 1 ha of forestland, different collection intensities caused different amounts of available biomass and, consequently, different amounts of carbon emissions from decomposition. When the functional unit was 1 ha of forestland, the type of biofuel product did not affect the biogenic CO₂ emission, biomass regrowth for compensation, or difference in carbon sequestration, because the biomass availability was the same under the same forest growth rate and collection intensity.

4.3. Limitations and Uncertainties

In the Carbon Budget 2021 [2], 11% of the anthropogenic CO₂ emissions originated from land use change and were mainly due to deforestation. In this study, the CCI of biofuel from land use change was considered by calculating the difference in carbon sequestration between the utilization scenario and the reference scenario. The values of the difference in carbon sequestration were negative in this study. However, land use change includes both direct and indirect land use changes. Indirect land use change emissions are difficult to quantify and were not included in the framework of this assessment.

Although the use of logging residues as biomass has positive impacts on climate change and reduces the risk of forest fire [48], there were also some negative impacts on other aspects, in consideration that logging residues play many important roles in forest ecosystems, such as habitats for microorganisms, sources of soil fertility, regulators of water resources, and maintainers of biodiversity [49,50,51]. Some compensatory management of the forest ecosystem, such as nitrogen fertilization, is needed after utilization, but this also increases the economic cost. Therefore, Hagenbo et al. [52] in the determination of land use strategy, a more comprehensive framework should be developed to deal with the tradeoffs among different impacts. Additionally, compared to biofuels produced from edible energy crops, biofuels produced from logging residues is not yet fully commercialized, so the production technology is not mature enough and the available data on the technical aspects are not of high enough quality and reliability [53].

This study focused on the carbon dynamics of forest types with varied growth rates. The integration of the Chapman-Richards function and the YASSO15 model can produce straightforward and reliable simulations without involving too many inputs. When more factors are involved, such as forest fire, human disturbance, and pests, process-based models should be chosen for accurate simulation, such as CENTURY [54], Triplex [55], and Forest Vegetation Simulator [56].

5. Conclusions

In this study, a new framework was implemented to assess the climate change impact of bioethanol and bio-diesel from logging residues, and the performance of this framework was analyzed under different forest growth rates and collection intensities of logging residues. According to the simulation, fossil fuel-derived GHG emissions, biogenic CO₂ emissions, and biomass regrowth for compensation were positive, and the sum of these three components was significantly higher than the climate change impacts from energy equivalent fossil fuels. The difference in carbon sequestration was negative because the use of logging residues can avoid carbon emissions from decomposition. The difference in carbon sequestration could offset the most positive CCI from the other three components and ensure a lower life-cycle CCI of biofuel in comparison with fossil fuels. The life-cycle CCI could even be negative in fast-growing forests with medium and high collection intensities of logging residues.

Above all, logging residue-derived bioethanol from fast-growing forests with high collection intensity could have a high CCI and a high mitigation effect. Fortunately, all scenarios estimated by the new LCA framework have a lower CCI than alternative fossil fuels. Collecting all available logging in the fast-growing forest to produce bioethanol is an optimal option for policymakers to mitigate climate change by replacing fossil fuels with bioenergy. However, from the perspective of land use strategy, the optimal option may not be attractive anymore.