1 Introduction

Climate change poses one of the most significant challenges to humanity today. The goal of limiting temperature rise to below 2℃, and more ambitiously below 1.5℃, as set by the Paris Agreement aims to mitigate the risks associated with global warming (Rogelj et al. 2016; Rogelj et al. 2018). China has pledged to enhance its nationally determined contributions, targeting a peak in carbon dioxide (CO2) emissions by 2030 and striving for carbon neutrality by 2060. To achieve these targets, urgent measures are needed to reduce CO2 emissions from high-emission sectors and increase CO2 removal from the atmosphere. According to the 2006 IPCC Guidelines for National Greenhouse Gas Inventories (IPCC 2006), the main sources of carbon emissions include the energy sector, industrial processes, agriculture, forestry and waste. Performing a carbon emission inventory on these sectors will provide insight into carbon balance and support efforts to mitigate temperature rise. This data will guide the deployment of CDR technologies suitable for local conditions.

Biochar, a carbon-rich product derived from the high-temperature pyrolysis of biomass, is frequently applied to soil to improve soil quality. Its inert nature makes it suitable for carbon sequestration (Lehmann 2007; Wu et al. 2019) and biochar is considered a promising CDR technology in China’s foreseeable future. Utilizing 73% of China’s crop residues for biochar production could lead to a cumulative reduction of 8620 Mt CO2e by 2050 (Yang et al. 2021b). Globally, the net CDR of biochar could offset 12% of CO2 emissions (Woolf et al. 2010). Moreover, utilizing biomass from bioenergy crops for biochar production can significantly contribute to achieving temperature control goals. Life cycle analysis (LCA) is a commonly used method for estimating the climate and environmental impacts of various CDR technologies (Terlouw et al. 2021). The CDR potential of biochar derived from crop residues or bioenergy crops has been systematically evaluated using LCA at scales ranging from the country to global level (Løvenskiold et al. 2022; Yang et al. 2021b).

Guangdong province is rich in agricultural and forestry biomass resources, with considerable potential in bioenergy crops biomass. Making full use of these biomass resources, particularly agricultural and forestry residues, can offer a sustainable alternative to fossil fuels, leading to reduced greenhouse gas and air pollutant emissions (Karan et al. 2023; Yang et al. 2021b). Bioenergy crops exhibit rapid growth and adapt well to relatively low-quality land conditions, making them suitable for cultivation on marginal lands such as abandoned cropland (Campbell et al. 2008; Nijsen et al. 2012). This approach helps prevent the encroachment on cropland, reduces competition with food crops and forests, and supports climate change mitigation efforts. Common bioenergy crops, including perennial grasses like miscanthus and switchgrass, as well as woody crops like eucalyptus, poplar and willow, have been extensively studied for their potential in climate mitigation (Gvein et al. 2023; Li et al. 2020).

Guangdong province is a major economic province in China, playing a pivotal role in the nation’s overall economic development. However, the rapid economic growth led to a significant rise in CO2 emissions and various environmental challenges. To effectively tackle these issues, conducting a comprehensive assessment of the carbon balance in Guangdong province is essential. Evaluating the biomass resource and the CDR potential of biomass-derived biochar becomes crucial in mitigating the province’s considerable carbon emissions. These evaluations can offer valuable theoretical insights and serve as references for government decision-making departments in developing climate mitigation strategies tailored to the unique challenges and opportunities of the region.

In this study, we conducted a comprehensive assessment of the land carbon balance across five key domains: (a) carbon sink from the land forest, (b) emissions from energy consumption, (c) emissions from industrial processes, (d) emissions from agricultural activities and (e) emissions from waste management. Our analysis covers the period from 2016 to 2021 in Guangdong province. Then, we assessed the biomass resource potential of collectable agricultural residues, forestry residues, and potential bioenergy crops cultivated on two types of marginal lands in Guangdong province, using a high-resolution (1km × 1km) biomass resource dataset (Wang et al. 2023). Finally, we quantified the CDR potential of biomass-derived biochar implementation in Guangdong province using LCA. Specifically, for bioenergy crops, we focus on five types: Miscanthus, Eucalyptus, Poplar, Switchgrass, and Willow. By analyzing and comparing the CDR potential of biochar derived from different biomass types, we assessed the capacity of biochar implementation to offset CO2 emissions in Guangdong province.

2 Material and methods

2.1 Study region

Guangdong province (20°13′–25°31′ N, 109°39′–117°18′ E) is located in southern China, bordering the South China Sea. The province comprises 21 prefecture-level cities (Fig. S1a) and is divided into four regions: The Pearl River Delta, Eastern Guangdong, Western Guangdong, and Northern Guangdong (Fig. S1b) according to the Guangdong statistical yearbook. Guangdong province has a total land area of 179,800 square kilometers, and the main land use types are forest and cropland (Fig. 1). The forest coverage rate in Guangdong province reached 58.74% in 2021. The region has a subtropical climate characterized by humid and rainy conditions. Over the period from 2000 to 2021, the annual mean temperature, precipitation, and annual mean sunshine hours were recorded at 22.3 ℃, 1652 mm, and 1767 hours, respectively (Fig. S2).

Fig. 1
figure 1

The land cover map of Guangdong province (resolution is 30m×30m) (Gong et al. 2019)

Full size image

2.2 Estimation of land carbon balance

We collected the activity data from five major sectors in Guangdong province: (a) land forestry, (b) energy consumption, (c) industrial processes, (d) agricultural activities, and (e) waste management. The calculation of greenhouse gas emissions within the specified land boundary followed the guidelines outlined in the Provincial Greenhouse Gas Inventory Compilation Guide (Trial) (http://www.cbcsd.org.cn//sjk/nengyuan/standard/home/; Table S1). It is important to mention that our calculations did not include carbon sinks from the ocean, such as shellfish carbon sinks. To ensure consistency for comparison and analysis in Guangdong province, methane (CH4) and nitrous oxide (N2O) emissions were converted to CO2e. Moreover, we conducted a comparative assessment of carbon emissions among four regions within the province.

2.2.1 Land forest carbon sink

We collected data on standing live timber volume (Vtot, m3) for forests in Guangdong province from 2015 to 2021, sourced from the Guangdong statistical yearbook. The land forest carbon sink (\({\text{C}}_{forest}\), Eq. [1]) was determined by assessing the changes in forest carbon stock between consecutive years. The estimation of forest carbon stock (\({\text{C}}_{\text{T}}\)) considered factors such as forest timber volume, the weighted averaged wood density (\(\overline{\text{SVD} }\), Mg DM m-3), a fixed biomass conversion factor for total biomass (BEFtot, ratio of forest biomass to timber volume), and a carbon conversion factor (\({F}_{c}\) = 0.5). It is important to note that our study did not account for carbon emissions from forest burning and decomposition.

$${\text{C}}_{forest,{T}_{i}}=\frac{{\text{C}}_{{T}_{i}}- {\text{C}}_{{T}_{i-1}}}{{\text{T}}_{\text{i}}-{\text{T}}_{\text{i}-1}} \times \frac{44}{12}$$
(1)
$${\text{C}}_{{T}_{i}}={V}_{tot}\times \overline{SVD}\times {BEF }_{tot}\times {F}_{c}$$
(2)

, where \({\text{C}}_{forest,{T}_{i}}\) (t CO2 yr-1) is the forest sink in year \({T}_{i}\). \({\text{C}}_{{T}_{i}}\) (t C yr-1) is the forest carbon stock in year \({T}_{i}\). The coefficient of 44/12 is used for converting carbon to CO2.

2.2.2 Energy consumption emissions

We obtained data on the terminal energy consumption of raw coal, oil products, electricity, and other energy sources for the past six years (2016-2021) in Guangdong province. Carbon emissions resulting from this energy consumption were estimated using the activity data and relevant emission factors. The emission factors for raw coal, oil products, and other energy sources were sourced from the Provincial Greenhouse Gas Inventory Compilation Guide (Trial) (Table S1). Specifically, the emission factor for oil products was calculated as an average of gasoline, kerosene, and diesel. The electricity emission factor was derived from values in region of South China (Table S1).

2.2.3 Industrial processes emissions

Emissions from industrial processes refer to those generated during the production of industrial products. It is important to note that CO2 emissions resulting from energy consumption in these industrial processes are not concluded here, as they have been accounted for in the energy consumption section (Sect. 2.2.2). Our data collection covered the total production of key industrial products in Guangdong province, including cement and steel. For steel production, where specific raw material usage amounts related to CO2 emissions (e.g., limestone, raw iron) were not available, we used previously established emission factor for the steel production process (0.213 t CO2 t-1 steel) by Li et al. (2016). The emission factors for cement and steel are detailed in Table S1.

2.2.4 Agricultural activities emissions

Emissions from agricultural activities encompass four primary components: CH4 emissions from rice fields, N2O emissions from cropland, CH4 emissions from animal intestinal fermentation, and greenhouse gas (i.e., CH4 and N2O) emissions from animal excrement management. In these activities, CO2 emissions are generally lower than CH4 and N2O. Moreover, CH4 and N2O have higher global warming potentials (GWP) at the 100-year scale, being 27 and 273 times that of CO2, respectively (IPCC 2022). Therefore, we only focused on CH4 and N2O in our carbon emission calculations from agricultural activities.

(1) For CH4 emissions from rice fields, we collected data on rice types (i.e., early rice and late rice) and the total sowing area in Guangdong province from 2016 to 2021. The emission factors for rice field activities in South China were used (Table S1).

(2) For N2O emissions from cropland, we calculated both direct and indirect N2O emissions. Direct emissions arise from the application of nitrogen fertilizer and manure to soil, as well as from straw returned to the field. In our calculations, we specifically focused on N2O emission from nitrogen fertilizer, using the available data on nitrogen fertilizer application (\({N}_{agr\_input}\)) in Guangdong province from 2016 to 2021. The emission factor for N2O in cropland in the Guangdong region was applied (Table S1). Indirect emissions include those resulting from atmospheric nitrogen deposition (\({C}_{dep}\)), as well as leaching and runoff (\({C}_{lea\_run}\)):

$${C}_{dep}={(N}_{exc}\times 0.2+{N}_{agr\_input}\times 0.1)\times {F}_{dep}\times \frac{44}{28}\times {GWP}_{N2O}$$
(3)
$${C}_{lea\_run}={N}_{agi\_input}\times 0.2\times {F}_{lea\_run}\times \frac{44}{28}\times {GWP}_{N2O}$$
(4)

, where \({N}_{exc}\) represents the nitrogen excretion from total animals, calculated using the animal numbers and N2O emission factors for each animal (Table S1). \({F}_{dep}\) and \({F}_{lea\_run}\) are the indirect N2O emission factors for atmospheric nitrogen deposition, and leaching and runoff, respectively (Table S1). The coefficient of 44/28 is used for converting nitrogen to N2O, and GWPN2O is the GWP of N2O ( = 273 at 100-year scale, IPCC 2022).

(3) For CH4 emissions from animal intestinal fermentation, we collected data on the numbers and types of animals in Guangdong province. The main animal types include yellow buffalo, cow, goat, pig and poultry. The corresponding CH4 emission factors for animal intestinal fermentation for each type are specified in Table S1. CH4 emissions are calculated by multiplying the number of animals in each category by their respective CH4 emission factors. These CH4 emissions were converted to CO2e using GWP of CH4 ( = 27 at 100-year scale, IPCC 2022) for comparison.

$${C}_{enteric}=\sum {(F}_{CH4\_enteric, i} \times {AP}_{i})\times {GWP}_{CH4}$$
(5)

, where Centeric represents the total CO2e emissions from all animals. \({F}_{CH4\_enteric, i}\) is the CH4 emission factor from the intestinal fermentation of the ith animal species. APi is the number of the ith animal species. GWPCH4 is the GWP of CH4 at 100-year scale.

(4) For CH4 and N2O emissions from animal excrement management, we used the emission factors specific to each animal type in South China (Table S1).

2.2.5 Waste management emissions

Emissions of CH4 may occur during wastewater treatment under anaerobic conditions, while the presence of nitrogen in wastewater can lead to N2O emissions. We collected data on solid waste volume, chemical oxygen demand (COD), and nitrogen content in wastewater for Guangdong province. The CO2 emission factors for solid waste, CH4 emission factor (average of domestic sewage and industrial wastewater), and N2O emission factor for wastewater are shown in Table S1.

2.3 Biomass resource potential

We assessed the biomass resources in Guangdong province using a detailed high-resolution (1 km × 1 km) biomass resource dataset for China (Wang et al. 2023). This dataset encompasses agricultural residues, forestry residues, and five bioenergy crops (Miscanthus, Eucalyptus, Poplar, Switchgrass, and Willow) cultivated on two types of marginal land: (1) marginal land I: identified as low productivity land based on land use types, and (2) marginal land II: identified as abandoned cropland based on cropland area changes from 1980 to 2018. The analysis also included the bioenergy crop types with maximum yield among the five types in each grid cell, referred to as the "best type".

The collectable agricultural and forestry residues are represented as grid data, and it is the total biomass amount in each 1 km2 grid. The grid data were allocated from total statistical data in each province based on forest and crop net primary productivity. For marginal lands, we assumed that the entire 1 km2 area is used for bioenergy cultivation. In summary, our assessment focused on agricultural residues, forestry residues, and potential bioenergy crops planted on marginal lands in Guangdong province. These various types of biomass were considered as feedstock for subsequent biochar production.

2.4 LCA of biochar

LCA is a widely used method for comprehensively evaluating the carbon mitigation potential, economic benefits, and environmental impacts of various processes. The carbon absorbed by plants through photosynthesis is reflected in their biomass. The biomass is then used to produce biochar, which fixes the carbon absorbed by plants and prevents the release of CO2 from biomass decomposition. To focus specifically on the carbon sequestration potential of biochar, the LCA of biochar begins from biomass collection, avoiding duplication of the carbon absorbed by plants. It is also consistent with many other studies of biochar LCA (Roberts et al. 2010; Yang et al. 2021a). The LCA boundary for biochar in Guangdong province is defined as the processes from feedstock supply to biochar application (Fig. S3). These processes include (a) feedstock collection, (b) transportation of feedstock and biochar, (c) biomass pyrolysis and biochar production, (d) application of biochar to cropland, (e) energy generation (e.g., biogas and bio-oils) from biochar production, (f) changes in soil organic carbon (SOC) with biochar addition, and (g) changes in N2O emissions and NPK fertilizer use efficiency resulting from biochar application to cropland.

In this study, we utilized agricultural and forestry residues, as well as the five bioenergy crops (i.e., Miscanthus, Eucalyptus, Poplar, Switchgrass, and Willow) cultivated on marginal lands in Guangdong province for biochar production. Byproducts generated from biomass pyrolysis, such as biogas, were assumed to replace carbon emissions from fossil fuel combustion. Biochar derived from different biomass resources was then transported to the nearest cropland and applied, resulting in varied effects on SOC and greenhouse gases like N2O. The addition of biochar resulted in an increase in SOC primarily due to the stable carbon content in biochar (Roberts et al. 2010). Moreover, applying biochar could lead to a 25% reduction in N2O emissions from croplands (Hammond et al. 2011). Because biochar contains nutrients that serve as fertilizers, its application reduces the need for chemical fertilizers, thus reducing greenhouse gas emissions associated with fertilizer use. Assumed parameters used in the LCA of biochar are provided in Table S2.

In this study, we assumed that biomass would undergo in-situ processing and pyrolysis using portable equipment, akin to the technology provided by Takachar (https://takachar.com/). Biochar, with its unique characteristics, is commonly used as a soil enhancer, offering several benefits such as improved soil quality, increased crop yield, and enhanced potential for soil carbon sequestration (Woolf et al. 2010). The growing endorsement of biochar in global agriculture highlights its significance as a pivotal strategy in mitigating climate change within agroecosystems (Bai et al. 2019; Zhou et al. 2017). Therefore, biochar derived from agricultural residues was returned to the in-situ cropland, while biochar derived from forestry residues and bioenergy crops cultivated on marginal lands was transported to the nearest cropland. To determine the shortest distance from the center point of each 1 km × 1 km grid site for forestry residues or bioenergy crops to the cultivated cropland, we used network analysis in Python. The road network data for Guangdong province were sourced and extracted from World Roads (https://esri.maps.arcgis.com/home/index.html). Considering the beneficial effects of biochar on improving soil quality, especially on relatively barren lands, and its ability to reduce carbon emissions during transport, we conducted a sensitivity test by assuming that biochar derived from forestry residues and bioenergy crops would be returned to the sites where the feedstocks were collected.

We allocated 24% of agricultural residues for poultry feed, 3% for industrial raw materials (Yang et al. 2010), and the remaining biomass was available for biochar production. For forestry residues, 50% (about the fraction of forest logging to all forestry residues in Guangdong, see Table S3) of the total amount was allocated for biochar production. Regarding bioenergy crops cultivated on marginal lands, the total biomass was considered available for biomass production, with a collective coefficient of 84% (Yang et al. 2010).

2.4.1 Scenario analysis

Considering the multiple utilization of residues in reality, we set five scenarios based on different assumptions on biomass availability for biochar production in Guangdong province:

  1. 1.

    The maximum scenario (S1) includes agricultural residues, forestry residues, and bioenergy crops cultivated  on two types of marginal lands;

  2. 2.

    The medium scenario (S2) includes agricultural residues, forestry residues, and bioenergy crops cultivated on  marginal land I;

  3. 3.

    The median scenario (S3) includes agricultural residues, forestry residues, and bioenergy crops cultivated on  marginal land II;

  4. 4.

    The residues-only scenario (S4) includes agricultural residues and forestry residues;

  5. 5.

    The minimum scenario (S5) includes only 11% of agricultural residues and 11% of forestry residues.

The S5 scenario reflects a near-reality situation because the actual amount of residues available for biochar production is very small (<1% of crop residues; Lu et al. 2022). A large portion of residues is utilized for various purposes such as fertilizer, feed, base, fuel, and raw materials in China (Wang et al. 2017; Huo et al. 2019). Therefore, we adopted a conservative estimation by assuming that only 11% of agricultural and forestry residues could be allocated for biochar production, based on the fraction of residues used for fuel production reported by Huo et al. (2019). Note that the maximum biomass was selected in cases where grid cells for different biomass types overlapped, such as abandoned cropland and forestry land.

2.4.2 Uncertainty analysis

We used the Monte Carlo simulation to evaluate the uncertainty of the CDR potential of biochar. This analysis considered the variability in input parameters, with parameter values sourced from current literature (Table S4). Random values for these parameters were generated with normal distribution. Through running 10000 iterations of the simulation, we calculated the mean and 95% confidence interval (CI) to determine the uncertainty range for the CDR potential of biochar.

3 Results

3.1 Land carbon balance assessment

The total carbon emissions in Guangdong province were significantly higher than the carbon sink, indicating that the province had not yet achieved carbon neutrality (Fig. 2). Among the five sectors contributing to carbon emissions, energy consumption accounted for the highest proportion at 83.8%, followed by industrial processes at 10.7%, agricultural activities at 3.2%, and waste management at 2.3% in 2021 (Fig. 2). Detailed emissions data for each sector are shown in Table 1.

Fig. 2
figure 2

The land carbon balance of Guangdong province from 2016 to 2021. Positive and negative values represent carbon sources and carbon sinks, respectively

Full size image
Table 1 The inventory of land carbon balance in Guangdong province (Tg CO2e yr-1)
Full size table

Between 2016 and 2021, land forests in Guangdong province acted as a carbon sink, absorbing approximately 32.42 Tg CO2 in 2021. During the same period, emissions from energy consumption consistently increased, reaching 802.6 Tg CO2e yr-1 in 2021 (Table 1), with emissions from electricity consumption contributing the most (Fig. S4a). Emissions from industrial processes also showed an upward trend, with cement and steel production processes emitting 91.49 Tg CO2e yr-1 and 10.89 Tg CO2e yr-1 in 2021, respectively (Table 1 and Fig. S4b). In agricultural activities, paddy fields accounted for the highest carbon emissions, followed by emissions from cropland, animal excrement management, and intestinal fermentation (Table 1). CH4 emissions from paddy fields remained relatively stable (Fig. S5a), while N2O emissions from cropland exhibited a decreasing trend annually, with direct emissions being approximately nine times higher than indirect emissions (Fig. S6a). Specifically, pigs contributed the most to emissions from excrement management (Figs. S7b-c and S8a), while water buffalo have the highest emissions from intestinal fermentation (Fig. S7a). Emissions from wastewater processing were higher than those from solid waste burning (Table 1 and Fig. S9).

Regionally, CH4 emissions from paddy fields exhibited spatial variations, with the highest emissions observed in Northern Guangdong (Fig. S5b). On the other hand, N2O emissions from cropland were the largest in Western Guangdong (Fig. S6b). Emissions from poultry excrement management were mainly concentrated in Western Guangdong and Northern Guangdong (Fig. S8b). Emissions from solid waste and wastewater management were higher in the Pearl River Delta than in other regions (Fig. S9b, d).

3.2 Biomass potential analysis

Based on high-resolution grid data, the total amounts of agricultural residues and forestry residues in Guangdong province were 12.29 Tg and 19.19 Tg, respectively (Table S3). The potential areas of marginal land I and marginal land II (i.e., abandoned cropland) in the province were estimated at 14,510 km2 and 22,294 km2, respectively (Fig. S10). Cultivating the best bioenergy crops on these two types of marginal lands had the potential to yield a maximum biomass of 21.36 Tg and 32.17 Tg, respectively (Fig. 3).

Fig. 3
figure 3

The distribution of biomass resources in Guangdong province with agricultural residues (a), forestry residues (b), best bioenergy crops yield on marginal land I (c) and marginal land II (i.e., abandoned cropland) (d). The pie charts embedded in (a) and (b) depict the proportions of seven types of crops and three sources of forestry residues, respectively. The bar charts in (c) and (d) display the potential yield of five bioenergy crops (i.e., Miscanthus [Mis.], Eucalypt [Euc.], Poplar [Pop.], Switchgrass [Swi.] and Willow [Wil.]) cultivated on the marginal land I and marginal land II in Guangdong province, respectively. Note that the best bioenergy crop type refers to the bioenergy crop type with the maximum yield among the five crop types in each grid cell

Full size image

Among agricultural residues, rice (75%) and peanuts (11%) were the most predominant crop types, followed by potatoes (8%) and maize (5%). Other crop types accounted for only 1% (Fig. 3 and Table S3). For forestry residues, the proportions from timber harvesting and processing, bamboo harvesting and processing, and forest logging were 34%, 18%, and 48%, respectively (Fig. 3 and Table S3). Among the bioenergy crop types, miscanthus and eucalyptus exhibited the highest biomass potential (Fig. 3). Details regarding other sources are shown in Table S3.

Spatially, agricultural residues were most abundant in North Guangdong, followed by the Pearl River Delta and West Guangdong region, with rice being the predominant crop type (Fig. 4a). Forest residues were primarily found in Northern Guangdong, followed by the Pearl River Delta region (Fig. 4b). The main sources of forestry residues included wood exploitation and processing, as well as forest logging (Fig. 4b). For bioenergy crops cultivated on marginal land I, the largest biomass potential occurred in Northern Guangdong (Fig. 4c). Regarding bioenergy crops cultivated on marginal land II (i.e., abandoned cropland), the region with the highest biomass potential was the Pearl River Delta, followed by Northern Guangdong (Fig. 4d). Eucalyptus emerged as the preferred bioenergy crop species cultivated on both types of marginal land (Fig. 4c, d). The distribution of biomass potential for the five bioenergy crops cultivated on the two types of marginal lands is shown in Figs. S11-S12.

Fig. 4
figure 4

The biomass resource potential in four regions of Guangdong province, including (a) agricultural residues, (b) forestry residues, (c) bioenergy crops on marginal land I and (d) bioenergy crops on marginal land II (i.e., abandoned cropland)

Full size image

3.3 LCA of biochar

3.3.1 CDR potential of biochar

The CDR potential of biochar in Guangdong province varies depending on the feedstock types. Biochar derived from the best bioenergy crops cultivated on marginal land II (i.e., abandoned cropland) showed the highest CDR potential at 31.93 Tg CO2e yr-1, followed by the best bioenergy crops cultivated on marginal land I (21.20 Tg CO2e yr-1), forestry residues (12.65 Tg CO2e yr-1) and agricultural residues (10.67 Tg CO2e yr-1) (Fig. 5). Among the five types of bioenergy crops, biochar derived from eucalyptus and miscanthus exhibited higher CDR potential, while poplar, switchgrass, and willow showed lower CDR potential (Fig. S13).

Fig. 5
figure 5

The total CDR potential of biochar derived from different feedstock of agricultural residues, forestry residues, best bioenergy crops on marginal land I and best bioenergy crops on marginal land II (i.e., abandoned cropland) in Guangdong province

Full size image

The main contributors to biochar CDR were the replacement of fossil fuel combustion with renewable energy (56.9%) and the increase in SOC (42.9%). A smaller proportion of CDR (0.26%) came from reduced fertilizer use and decreased agricultural N2O emissions (Fig. 5). The main emission sources were from the feedstock collection and processing processes, followed by the biochar spreading process. Emissions from feedstock and biochar transportation were relatively small. The average distances from marginal land I, marginal land II, and forest sites to the nearest cropland were 6.60 km, 5.50 km, and 6.42 km, respectively (Fig. S14), with marginal land I and forest residues sites being farther than marginal land II. Because distance has little impact on carbon emissions, applying biochar derived from forestry residues and bioenergy crops directly to the soil at their in situ feedstock sites showed a similar CDR potential (Fig. S15) to that applying biochar to the nearest cropland (Fig. 5; Sect. 2.4).

The spatial distribution of biochar CDR potential was mainly influenced by the type of feedstocks used (Figs. 4 and 6). The highest CDR potential of biochar, derived from agricultural and forestry residues, as well as the best bioenergy crops cultivated on marginal land I, was concentrated in Northern Guangdong. The CDR potential of biochar from the best bioenergy crops cultivated on marginal land II was distributed in the Pearl River Delta, Northern Guangdong, and Western Guangdong regions (Fig. 6). Among the five bioenergy crop types, eucalyptus and miscanthus were the major contributors in these four regions (Fig. S16). The spatial distribution of the CDR potential of biochar from these five bioenergy crops is depicted in Figs. S17-18.

Fig. 6
figure 6

The spatial CDR potential of biochar derived from (a) agricultural residues, (b) forestry residues, (c) best bioenergy crops cultivated on marginal land I, and (d) best bioenergy crops cultivated on marginal land II (i.e., abandoned cropland) in Guangdong province. (e) shows the CDR potential of biochar from different feedstock in four regions of Guangdong province. (f) shows the CDR potential of biochar under five scenarios of biomass resource utilization (S1–S5) in these four regions of Guangdong province

Full size image

3.3.2 Scenario analysis

The maximum biomass potential for the five scenarios of biomass resource utilization was observed in northern Guangdong (Fig. 6), with the highest CDR potential found in the maximum scenario S1 (agricultural residues, forestry residues, and best bioenergy crop yields on two types of marginal land) (Fig. 6f). In scenario S1, the maximum CDR potential of biochar for the entire Guangdong province reached 84.30 Tg CO2e yr-1 (Fig. 6f), which can offset 9.11% of net carbon emissions (Table 1) in Guangdong province. In the residue-only scenario of S4, the CDR potential of biochar can offset 4.29% of net carbon emissions. However, in the minimum resource utilization scenario (S5), the CDR potential of biochar only offset 0.052% of net carbon emissions in the province, highlighting the significance of cultivating bioenergy crops on marginal lands as a promising option for CDR in the future.

3.3.3 Uncertainty analysis

To evaluate the uncertainty of biochar CDR potential in Guangdong province, the Monte Carlo simulation was conducted by iterating main biochar-related parameters (Table S4; Sect. 2.4). Our results showed that the CDR potential of biochar in Guangdong province was 9.61 (95% CI: 7.85–11.77 Tg CO2 yr-1), 10.21 (95% CI: 8.25–12.53 Tg CO2 yr-1), 19.63 (95% CI: 16.08–24.00 Tg CO2 yr-1) and 29.44 (95% CI: 24.04–35.99 Tg CO2 yr-1) for four feedstock types of agricultural residues, forestry residues, bioenergy crops on marginal land I and marginal land II, respectively.

4 Discussion

4.1 Comparison of land carbon balance with other studies in Guangdong province

The estimation of forest carbon sinks involves various factors and assumptions, leading to potential variations based on different methodologies and data sources. According to the estimation of forest carbon stock change per area in China from 2001 to 2010 (0.62 Mg C ha-1 yr-1) by Fang et al. (2018), the forest carbon sink in Guangdong province is estimated to be 23.98 Tg CO2 yr-1 based on the Guangdong’s forest area (1054.7 × 104 ha) in 2021. Another estimation reported by Piao et al. (2009) (land carbon sink magnitude in China: 20–27 gC m-2 yr-1) suggests a range of 7.73–10.44 Tg CO2 yr-1 for the forest carbon sink in Guangdong province. In comparison, the forest carbon sink estimated in this study (20.29–37.23 Tg CO2 yr-1; Table 1) aligns closely with that reported by Fang et al. (2018) but is higher compared to Piao et al. (2009). It could be attributed to an increase in forest area, potentially resulting in the forest carbon sink in Guangdong province surpassing the average level observed in China during the 1980s and 1990s by Piao et al. (2009). This disparity could also stem from the lack of site-specific parameters, resulting in lacking spatial heterogeneity and representation. The constant parameters of wood density, biomass, and carbon conversion factors applied across the entire region (Sect. 2.2.1) might contribute to an overestimation of the forest carbon sink in Guangdong province. Our estimation of the forest carbon sink in Guangdong province is also higher than the 16.0 Tg CO2 reported by Lu et al. (2024) for 2018. The difference could be attributed to the exclusion of emissions from forest transformation processes such as burning or decomposition in our study. In addition, our estimation (Table 1) is comparable to the results reported by Lu et al. (2024) for CO2 emissions from energy consumption (732.2 Tg CO2), industrial processes (63.2 Tg CO2), agricultural activities (43.3 Tg CO2), and waste management (30.1 Tg CO2) in 2018, indicating the reasonableness of our estimation regarding the land balance of Guangdong province.

4.2 Uncertainty of LCA for biochar

The CDR potential of biochar can vary significantly depending on assumptions regarding pyrolysis processes (e.g., fast/slow pyrolysis), parameters, boundaries, and feedstock availability (Terlouw et al. 2021). For example, Yang et al. (2021a) estimated that about 8.0 Tg CO2 could be reduced in 2014 through biochar derived from available crop residues in Guangdong province, which is lower than our study’s estimation of 10.67 Tg CO2e from crop residues (Fig. 5). This discrepancy may arise from the availability of crop residues used for biochar production and other factors such as different feedstock types, various biochar properties, and diverse pyrolysis technologies (Deng et al. 2024; Lu et al. 2022). Our estimation of the uncertainty range for biochar CDR from agricultural residues was approximately 7.85–11.77 Tg CO2 yr-1 (about 0.87–1.31 t CO2e yr-1 t-1 agricultural residue), which aligns with findings by Yang et al. (2021a) (> 0.92 t CO2e t-1 biomass), Robert et al. (2010) (0.86 t CO2e t-1 stover), and Hammond et al. (2011) (0.7–1.3 t CO2e t-1 various residues). The consistency indicates that the framework of our biochar LCA is robust and reasonable, providing a convincing estimation of the CDR potential of biochar and its contribution to offsetting carbon emissions in Guangdong province.

This study considers agricultural and forestry residues sourced from high-resolution grid data (1 km × 1 km), relying on provincial statistics. However, the yield of bioenergy crops is derived from globally upscaled data using a random forest model with a coarse resolution (0.5° × 0.5°) (Li et al. 2020), potentially lacking spatial variability at smaller regional scales within Guangdong province. In addition, we only assessed five common types of bioenergy crops, whereas there might be tree species better suited for cultivation on marginal lands in the province. Therefore, the CDR potential of biochar could be further refined by incorporating site-specific data on bioenergy crop yields and considering additional local tree species suitable for the region.

The LCA of biochar conducted in this study relies on parameters sourced from various literature and assumed fixed values (Table S2). However, specific parameterization and spatialization for certain variables are crucial to reflect local conditions accurately in Guangdong province. For instance, the wide range of N2O emissions reported by Cayuela et al. (2015) may introduce uncertainties in the carbon balance assessment. CH4 emissions in rice paddies with biochar addition also vary with soil conditions such as soil moisture and pH (Jeffery et al. 2017). These factors were not considered in the study’s assessment of biochar's CDR potential due to significant uncertainties in these processes. Future studies should focus on the mechanisms of biochar addition on soil greenhouse gas emissions to minimize these uncertainties.

4.3 Potential technologies to achieve carbon neutrality in Guangdong

Our findings indicate that relying solely on biochar technology may not be adequate to achieve carbon neutrality in Guangdong province, with a potential offset of only 9.11% of net carbon emissions (Section 3.3.2). Urgent mitigation efforts are required to balance the carbon emissions in the region. Given the significant contribution of energy consumption to carbon emissions (Fig. 2), a primary emphasis on energy transformation and the adoption of clean energy alternatives is imperative. Industries, especially those in sectors like cement production, can play a pivotal role by implementing strategies such as enhancing energy efficiency, utilizing low-carbon fuels (Tan et al. 2022), and embracing innovative technologies like Direct Air Capture of CO2 with Storage (DACCS). Furthermore, optimizing and efficiently utilizing cement and concrete (Watari et al. 2022) will be a crucial strategy in advancing towards net-zero emissions.

On the other hand, increasing carbon sinks is equally important. Our analysis of forest age using remote sensing data from Zhang et al. (2017) revealed that the average forest age in Guangdong province is 17.7 years (ranges: 2–45 years; Fig. S19b). Yu et al. (2014) suggested that younger trees have a greater capacity for carbon sequestration. Therefore, prioritizing afforestation/reforestation (AR) programs that focus on planting younger trees, especially fast-growing bioenergy crops on marginal lands, can significantly enhance carbon sequestration during their growth phase. Making full use of the abundant biomass resources in Guangdong province, biomass-based CDR technologies like Bioenergy with carbon capture and storage (BECCS) will hold significant potential for carbon removal (Fan et al. 2021; Gvein et al. 2023). Combining strategies that combine BECCS with biochar could also play a pivotal role in mitigating carbon emissions in Guangdong province.

Guangdong province encompasses extensive marine areas spanning approximately 419,000 km2, including sizable mudflats covering about 1800 km2. Marine ecosystems hold the potential for significant negative carbon emissions through mechanisms like the microbial pump (Jiao et al. 2010). Developing these marine carbon sinks presents a valuable strategy for offsetting carbon emissions in Guangdong province (Liu et al. 2022; Wang et al. 2021). Therefore, exploring and evaluating ocean-based CDR technologies such as ocean fertilization (OF) could significantly contribute to climate change mitigation efforts. However, it is essential to assess the trade-offs among CDR potential, environmental risk, and economic viability associated with large-scale deployment of diverse land- or ocean-based CDR technologies in the future (Minx et al. 2018).

5 Conclusion

Our study provided a comprehensive inventory of the land carbon balance across five sectors and assessed the CDR potential of biochar implementation in Guangdong province. It offers technical guidance for deploying biochar and other CDR technologies in the region. Additionally, deploying other biomass-based CDR technologies like BECCS and evaluating the combined CDR potential of biochar and BECCS can help mitigate the significant CO2 emissions from energy and industry processes. Moreover, the additional demand effects of these technologies (e.g., large water consumption) and their potential feedback effects on climate should be taken seriously for future regional or national deployment.