Carbon reduction and water saving potentials for growing corrugated boxes for express delivery services in China

Significance

Courier over-packaging and disposal are causing serious environmental issues. Eighty percent of global express delivery growth happened in China, where the corrugated packaging demand increased 90-fold during the past decade. However, the existing knowledge on environmental impacts of producing and consuming corrugated boxes is confined to only life-cycle carbon emissions at 1-y snapshot. Facing the changing global supply chains of water-intensive raw materials and increasing rate of packaging recycling, here we estimate the inter-annual trends and drivers in cradle-to-grave life-cycle carbon emissions, water quantity and quality appropriation of corrugated boxes used in China over 2007-2021, as well as mitigation potentials for 2035. The results provide key information towards green express delivery industries in both China and the globe.

Abstract

Corrugated packaging for express grew by 90 times to 16.5 Mt y⁻¹ in China, where 81% of recent global express delivery growth occurred. However, the environmental impacts of production, usage, disposal, and recycling of corrugated boxes under the entire supply chain remain unclear. Here, we estimate the magnitudes, drivers, and mitigation potentials of cradle-to-grave life-cycle carbon footprint (CF) and three colors of water footprints (WFs) for corrugated cardboard packaging in China. Over 2007 to 2021, CF, blue and gray WFs per unit package decreased by 45%, 60%, and 84%, respectively, while green WF increased by 23% with growing imports of virgin pulp and China’s waste ban. National total CF and WFs were 21 to 102 folded with the scale effects. Only a combination of the supply chain reconstruction, lighter single-piece packaging, and increased recycling rate can possibly reduce the environmental footprints by 24 to 44% by 2035.

Courier over-packaging and disposal are causing serious carbon emissions and water degradations from the associated globalized supply chains worldwide, challenging the packaging industry to grow within the sustainable limits of environmental impacts (1–4). Corrugated boxes are the worldwide most common packaging material for express delivery industries and treated as an eco-friendly succedaneum of plastic packages as being easy to recycle, renewable, and biodegradable (5–7). The global demand of corrugated boxes has been five-folded over the last decade, with following expected 2.6% more demand per year (8). The complex cradle-to-grave life cycle of a corrugated box encompasses the raw material stage (RMS) for preparing the pulp, the box-making stage (BMS), the usage stage (USS), and the end-of-life stage (ELS) (Fig. 1). In the ELS, the disposal of a corrugated box can be recycling, incineration, or landfill. The water-intensive agroforestry, the energy-, and pollution-intensive pulp and paper industries are involved in the life cycle of a corrugated box (9–11). The chemically intensive pulp and paper industry is one of the world’s biggest contributors of freshwater consumption, energy usage (~4% of world energy), and pollution (12). The corrugated box for express delivery is treated as a typical circular bioeconomic product, which comprises use and recycling of biological resources and waste (e.g., wood and crop straw for pulp), aiming for high resource efficiency (13). The life-cycle environmental footprint accounting has been recognized as an efficient approach to measure the environmental consequences from production, trade, and consumption along the life cycle (14).

Fig. 1.

The system boundary of the current analysis showing the life cycle of a corrugated box. The icons in the figure were reproduced from the Noun Project under a Creative Commons license CC BY 4.0.

Existing literatures on the environmental impacts and footprints of booming corrugated packaging for express delivery focus mostly on energy consumption, greenhouse gas emissions, or carbon footprints (CFs) (15–19). It has been claimed that packaging paper accounted for over 40% of global total greenhouse gas emissions in papermaking (15). The increasingly frequent bilateral burden shifts of the environmental impacts embedded in trade at the RMS have been the focal points (20, 21). However, the intensive green, blue, and gray water footprints (WFs) and the burden shifts of water-related environmental impacts along the life cycle of the corrugated boxes are still overlooked (9, 10). Along the life cycle of corrugated boxes, green WF refers to the rainwater consumed for growing trees and crops, which is the largest part of life-cycle water consumption (22, 23); blue WF refers to the surface and ground water consumption at each stage; gray WF refers to the volume of freshwater required to dilute pollutants emitted to water (24). Limited information is available yet on green and gray WFs of a corrugated box (9). The only record of the life-cycle blue WF of corrugated boxboard produced in China was 40 to 50 m³/t (11).

Over the past decade, China contributes 81% of global express delivery growth. The usage of corrugated cartons (~16.5 Mt/y) in China accounted for 9% of the global total by 2021 and was 90 times more than that in 2007 (SI Appendix, Fig. S1). Packaging in China’s express delivery industry generated carbon emissions of 13.2 Mt CO₂eq, where corrugated boxes account for over 80% (25). Owing to the implementation of the “waste ban” in 2017 (26), the supply chain structure of corrugated boxes production changed as China reported reduced foreign waste paper raw material imports, which has significantly impacted the global waste paper recycling patterns (21). The specific environmental potential benefits of circular economy efforts (e.g., more recycling), especially for paper-based products, which are influenced by specific raw material structure, production technologies, and energy use structures have been a focal point for recent research studies (27, 28). However, most attention in previous studies has been paid to 1-y snapshot of the environmental impacts of corrugated boxes or related paper products with different raw materials (29–31), different production technologies (32–35), and different life-cycle stages (36–38). The latest and most comprehensive inter-annual CF accounting for global paper and pulp industry over 1961 to 2019 (15) and CF mitigation scenarios specifically for China’s express delivery sector (39) still ignores the impacts of international waste paper trade, which is the fundamental material for producing corrugated boxes. Lack of understanding on effects of the changing global supply chains and temporal dimensions hinders the monitoring of the environmental burden and mitigation potential from a systematic perspective under a circular economy, as well as seeking more optimized green development pathways both from supply-side and consumption-side for corrugated boxes used in China’s express delivery industry.

To address the abovementioned research gaps, this study estimates the magnitudes and drivers of interannual developments of stage-by-stage cradle-to-grave life-cycle CF and WFs of the corrugated cardboard packaging for China’s express delivery industry over 2007 to 2021, which involves 26 relevant material export countries. We then performed a decomposition analysis to reveal the drivers of CF and WFs changes and a scenario analysis for mitigation pathways and potentials of both CF and WFs. There are two features of this analytic system structure. First, total environmental footprints comprise emissions and resource use on domestic territory plus those associated with imports in the consumption-based accounting. Second, the use of a life-cycle mechanism, both the environmental debit of the waste paper as raw material and the environmental credit of recycling the corrugated boxes after use by the express delivery industry are considered based on the fiber cycle numbers under one open-loop recycling system. The accounting of the environmental footprint at each stage considers the direct and indirect environmental footprint, i.e., the environmental footprint generated directly by the processes in each stage and the environmental footprint generated by the production of various products (e.g., electricity, chemicals) to which it contributes, respectively. The current analysis clarifies the recent trends in carbon emissions, water quantity and quality appropriation, and corresponding mitigation potentials toward a sustainable life cycle of corrugated boxes from a global supply chain perspective, thus providing key information to support policies achieving green pathways for Chinese as well as the global express delivery services.

Results

The Carbon and Water Footprints of Unit Mass of Corrugated Boxes.

For unit mass of corrugated cartons for express packaging in China by 2021, the CF, gray WF, blue WF, and green WF over its life cycle were 2.8 t CO₂eq/t, 370 m³/t, 23 m³/t, and 313 m³/t, respectively. Each of the environmental footprints has diverse compositions in terms of contributions by life-cycle stages and sources (Fig. 2). For example, energy input in box making stage contributed 64% of the CF, while only 5% of the blue WF. The unit CF (Fig. 2A) and gray WF (Fig. 2B) decreased significantly, by 45% and 84%, respectively. Smaller CF and gray WF are attributed to the production efficiency improvement, particularly the less energy consumption and chemical oxygen demand emissions of China’s pulp and paper industry since 2006 (25) (SI Appendix, Table S12). Although the unit blue WF decreased by 59% by 2017 (Fig. 2C), we found the unit blue WF later had stagnated from 2017 and increase slightly by 9% from 2017 to 2020 (Fig. 2C). The unit green WF increase by 23% during 2007 to 2021(Fig. 2D). The increase in unit blue and green WFs is dominated by the growing proportion from 46 to 55% of imported wood pulp in the virgin pulp from other countries (SI Appendix, Fig. S3), where the unit green and blue WFs of wood production are significantly higher than the Chinese local levels (SI Appendix, Fig. S6). In addition, we find that upon replacing virgin fiber production by more environmentally efficient recycled waste paper, the recycling of corrugated boxes had non-negligible environmental credits by mitigating both CF and WFs, by 11 to 17% and 18 to 95%, respectively (Fig. 2).

Fig. 2.

Magnitudes, composition, and inter-annual developments of (A) CF, (B) gray WF, (C) blue WF, and (D) green WF of full life-cycle per tonne corrugated cardboard in China over the period 2007 to 2021. The dot lines in the middle show the trends of unit environmental footprints. Different color areas represent the environmental footprints of different intermediate products or processes. In the Sankey diagrams, columns A, B, and C represent different environmental footprint sources, intermediate products or processes, and life-cycle stages, respectively. We only show sources greater than 1% of the total environmental footprints (see SI Appendix, Fig. S17 for the full information).

The Booming Total Carbon and Water Footprints of Corrugated Boxes in China.

Despite the presence of reducing CF, blue and gray WFs per unit corrugated case, the 90-time jump in the number of corrugated boxes used for express delivery over the same period translates to significantly higher rates of expansion in all the resultant national total environmental footprints. The national total green WF of corrugated cases grew the most, by 111 folds (Fig. 3D), followed by the CF (~50-folded, Fig. 3A), blue WF (~37-folded, Fig. 3C), and gray WF (~15-folded, Fig. 3B). Through imports of virgin pulp and waste paper for corrugated cardboard production, the embedded imported blue and green WFs accounted for 67% and 101% in national totals, respectively. The corrugated recycling led to negative domestic green WFs from 2012 to 2017, showing the environmental credit effect (Fig. 3D). Since the implementation of the import restriction policy of waste paper raw materials by China’s government from 2017 (26), the external ratios (i.e., the proportion of imported environmental footprints) in all the environmental footprints have been reduced visibly at different degrees, with the highest reduction rate of 96% observed for gray WF (Fig. 3B).

Fig. 3.

Growing national total (A) CF, (B) gray WF, (C) blue WF, and (D) green WF of corrugated boxes used for express delivery in China over 2007 to 2021.

Drivers of Growing National Carbon and Water Footprints of Corrugated Boxes.

We deconstructed contributions of key drivers for each stage (RMS, BMS, and ELS) along the life cycle of the corrugated cases for express delivery in China in terms of the effects of the scale, intensity, and structure on inter-annual developments of each considered environmental footprint over the study period (Materials and Methods). Results clearly show that the scale effects in RMS and BMS defined the huge growth in environmental footprints of China’s corrugated boxes as the positive drivers, with greater impacts on CF and gray WF than blue and green WFs, with limited offsetting ranging from 36 to 53% by corrugated box waste disposal in the ELS (Fig. 4). The 44-fold increase in corrugated box production contributed 33-fold and 65-fold increases in CF and blue WF, respectively (Fig. 4 A and C). This verifies the phenomenon shown in the last section that the reduction rate of environmental impacts by unit corrugated cardboard was far behind the growing rate of production and consumption scales within the country. The effects of scale drivers were higher in the RMS, where the contribution of scale effects to growth in green WF was 30 times that in the corresponding BMS (Fig. 4D).

Fig. 4.

Contribution of the scale, intensity, and structure effects to the booming national total (A) CF, (B) gray WF, (C) blue WF, and (D) green WF of corrugated boxes used for express delivery in China for 2007 to 2021.

In most cases, the drivers in terms of intensity effects were negative to worsening environment except for the green WF between the years 2007 and 2014 (Fig. 4D), as the higher green WF embedded in increasing imported wood pulp from Canada instead of the previous imports from Chile (SI Appendix, Fig. S3). Meanwhile, the drivers regarding structural effects were negative for emission footprints (i.e., CF and gray WF) in the RMS (Fig. 4 A and B), whereas positive for resource footprints (i.e., blue and green WFs) (Fig. 4 C and D).

Low-Carbon and Water-Saving Pathways of Corrugated Delivery Packaging in China.

In the face of the inexorable increasing trend of corrugated delivery packages toward the peak level around 2035 (39), setting green development pathways to minimize and mitigate environmental impacts is essential to meeting China’s commitment to both carbon neutrality and relative sustainable development goals (SDGs). We examined responses and mitigation potential for the considered emission and resource footprints related to corrugated cases in three action scenarios (S1-S3) from production perspective, two action scenarios (S4-S5) from the consumption perspective, one action scenario (S6) from the post-consumer perspective, and three combination scenarios (S7, S8, and S9) for 2035 (Fig. 5). Changing consumer habits by reducing the single delivery packaging by 2.5%/y (S4) is effective and has positive impact compared to actions in the production or post-consumer sectors on not only putting peaks of each environmental footprint forward but also mitigating all the tested environmental footprints by 20% compared to the reference level of maintaining current trends (S0) (Fig. 5). We found that only taking measures in improving productivities in terms of resource intensity (S1), wood pulp structure (S2), or energy structure (S3) in the production side will lead to trade-offs between reducing emission footprints and enhancing resource footprints. A combination scenario from the production side (S7) reduces the emission footprints by 16 to 30%, while increasing the resource footprint by 5 to 13%. If all the countermeasures across S1-S6 were combined, in the optimized S9, the tested environmental footprints related to corrugated packaging in China peaked by 2030 and were reduced by 24% (in green WF) to 44% (in gray WF) by 2035, compared to the levels in S0.

Fig. 5.

Environmental footprint reduction potentials of corrugated cartons in the express delivery industry over 2022 to 2035 and the corresponding peak years. The inter-annual trends in magnitudes of each environmental footprint under scenarios can be seen in SI Appendix, Fig. S21.

Discussion

Enhancing Synergies and Reducing Trade-Offs between Carbon and Water Footprints.

Information on past developments is essential to the development of future strategies toward the establishment of green packaging for express delivery services in China. Interannual developments in typical emission footprints (i.e., CF and gray WF) and resource footprints (40) (i.e., blue and green WFs) of the life cycle of express packaging production, consumption, and post-consumer disposal in China during 2007 to 2021 from a global supply chain perspective are estimated incorporating the open-loop recycling in the environmental footprint accounting framework. All the environmental footprints in the national total increased by 15 to 110 times owing to the drastic increase in usage, despite the significant reduction in the emission footprints of unit mass of corrugated cartons via various measures such as the implementation of a series of Cleaner Production Acts and further tightening environmental standards (41, 42) in terms of energy saving and water pollution control to achieve improvements in resource productivity and eco-efficiency. Trade-offs also show that compared with the decreasing emission footprints of the unit corrugated box, the trends of recently increasing corresponding resource footprints owing to additional imports of water-intensive foreign wood pulps implied a more worrying situation. As a critical and limited resource (10), green water is often disregarded in previous studies. Traditional independent focus on either GHG emissions or water pollutions of express delivery services is apparently insufficient. Our study indicates the possible risk of government policy or industry practice increasing water use while reducing carbon emissions in the express delivery industry. Multi-environmental footprint assessment is crucial for a green and sustainable express delivery industry aimed at enhancing synergies while reducing trade-offs among different kinds of environmental concerns.

Shared Responsibility Is Essential: Social and Economic Implementations.

Scenario analysis shows that except energy structure adjustment (S2) for the CF, the extent to which a single (S1, S2, and S3) or multiple measures from the production and post-consumer sides can reduce environmental impact is limited, thus making it difficult for environmental impacts to reach the peak value before 2030. Consumers determine the production behavior of producers and are the most important drivers of environmental burden. Post-consumer waste disposal also affects the environmental burden throughout the life cycle of corrugated boxes. Therefore, the green development of the express industry requires the concerted consideration of the shared and tripartite responsibilities from the government, e-commerce platforms, express service companies, packaging manufacturers, and consumers.

The current analysis sheds lights on the major contribution of consumer and post-consumer sides to mitigate environmental impacts of corrugated packaging for express. Social and economic efforts are apparently crucial for improvements in consumer and post-consumer sides. As an efficient way to help consumers understand the relative pieces of knowledge and the reasons of reuse or recycling of package materials rather than a general awareness of sustainable developments, a conspicuous label of environmental impacts, mitigation potentials as well as recommended consumption behaviors on express cartons can be prepared by producers. More importantly, the infrastructure needs to be enhanced and expanded for classifications and recycling of used corrugated packages. Economically, the core is minimizing the cost or maximizing the benefits while diminishing the visible negative environmental impacts along the life cycle of a corrugated box. It cannot be addressed without government subsidies or price leverages with the increasing costs when upgrading technologies in production (i.e., scenarios S1, S2, and S3) or applying novel entities to optimize logistics (i.e., S5) (43). In addition, the government subsidies are also necessary to replacing alternative packaging materials which are much more environmentally friendly but of higher cost than corrugated packages (2). We compare the environmental impacts between the same weight of reusable polypropylene plastic boxes and corrugated cartoons (SI Appendix, Fig. S24). The CF of unit mass of reusable polypropylene plastic packages is 2.6 times that of corrugated packages, whereas the blue WF is 18% smaller and gray WF of reusable polypropylene plastic packages is nearly null. However, the plastic packages of higher costs can be directly reused for 50 to 70 times (44). Much more environmental mitigation potentials can be achieved if just 25% of corrugated boxes were replaced by 50-time reused plastic packages in the tested scenarios (SI Appendix, Fig. S26).

Political Implementations and Countermeasures.

By comparing the environmental footprint changes of China’s express industry in the future with single-party and multi-party efforts, our study strongly recommends multi-end multi-party collaborative efforts to jointly promote the greening construction process of China’s express packaging, with more mature and comprehensive political countermeasures. For the production side, the monitoring and recording system of multiple environmental impacts at each chain and stage is urgent to be built and kept updated with corresponding evaluation standards. Policies should be further established for enhancing the financial support to innovative packaging initiatives of enterprises to accelerate the large-scale promotion of new recycling packaging materials. Currently, only certain large enterprises in China have conducted relative research studies; for example, the SF Express Company launched the “π-box.” Meanwhile, regarding the “environmental, social, and governance (ESP)” reporting by express enterprises, formulating unified standards and regulars for transparency increasement and obligation improvements are highly recommended. Since 2002, there have been regulations and standards in China to promote cleaner production with less carbon emissions and water usage and pollutions in paper industry, but little on improving consumer behaviors and post-consumer managements (SI Appendix, Table S17).

For the consumer side, although consumers are willing to participate in green packaging initiatives, their knowledge on specific environmental information is limited (45, 46). Several regulations to reduce over-packaging and increase public awareness of environmental impacts mitigation through packaging can be formulated. For instance, adding radio button group for recycled and reused packaging service for consumers on E-commerce platforms; setting strict classification standards for packaging size according not only to the characteristics of the goods but also to multiple associated environmental impacts. For the post-consumer side, a responsibility-sharing mechanism or benefit-sharing mechanism for different entities in the express business should be established. For example, the government can implement the Extended Producer Responsibility (EPR) policy (47) for e-commerce platforms, explore the inclusion of express packaging recycling within the scope of e-commerce services, encourage express enterprises to opt for collection and delivery outlets, self-service collection cabinets, and self-service pick-up sites and support the construction of express packaging classification and recycling facilities to achieve waste-efficient recycling of express packaging. Covering all the stages of the life cycle of typical packaging materials, fast construction of comprehensive environmental impact accounting and assessment standards and databases at product, corporation, and different administrative regional levels are crucial and highly recommended.

Limitations and Future Directions.

Limitations in the current analysis represent future research directions. First, the current estimation of carbon reduction and water-saving potentials for growing corrugated boxes in China is at national level. Given the data availability, spatial differences as resulted from varied energy structures, productivities at each life-cycle stage due to possible diverse technologies, as well as CF, WF, or other types of environmental footprint of unit raw materials cannot be accounted at present. More detailed measurements, records and surveys for energy use, material flows, environmental impacts, resource productivities at corporation and sub-national regional levels are highly recommended. Second, the necessity and importance of multiple environmental footprint assessment and associated trade-off management are highlighted in this analysis, but with a certain study case for corrugated packaging in China. Although an efficient multi-environmental-footprint accounting and mitigation potential assessment framework is shown, a more comprehensive picture is to be drawn with consideration of alternative packaging materials (e.g., bioplastics) and other types of environmental impacts including air pollution, land use, or ecotoxicity (48). Last but not least, facing challenges in achieving green packaging and express in the context of circular economy, environmental analysis is just the fundamental steps. Corresponding economic assessments, or even behavioural studies are crucial to generate wisely environmental, social, and economic countermeasures toward relative SDGs at different stages and levels.

Materials and Methods

System Boundary.

The corrugated boxes life-cycle process is a typical circular bioeconomic system and has a relatively long production supply chain. We adopted an open-loop recycling allocation method (49) modified from ISO 14067 (50) in the bottom-up consumption-based environmental footprint accounting framework. Fig. 1 illustrates the system boundary which comprises four main stages: the RMS, the BMS, the USS, and ELS. According to the Intergovernmental Panel on Climate Change guidelines (51), carbon emissions from biogenic sources are not included in the total carbon emissions. In this study, biomass is assumed to be fully combusted to be carbon neutral (CO₂ neutral), and CO₂ from biomass combustion in pulp and paper and waste incineration processes is excluded from the total carbon emissions of the system, which corresponds to carbon sink from crop residues and wood production was also not considered (52). The anaerobic digestion of biomass pollutants in the paper industry wastewater treatment results in CH₄ emissions that are included in this study.

Carbon and Water Footprints Calculation.

The CF and WFs of corrugated box packaging in the express delivery industry comprise four life-cycle stages including the RMS, BMS, USS, and ELS. The total environmental footprints (CF and three kinds of WFs) were calculated by multiplying the consumption of corrugated boxes by the corresponding unit environmental footprints. The unit environmental footprints of each stage were calculated by adding the unit direct footprint to the unit indirect footprint.

$${\mathit{EF}}_i=\left({\mathit{EF}}_{i,RS}+{\mathit{EF}}_{i,BS}+{\mathit{EF}}_{i,ES}\right)\times P,$$ [1]

$${\mathit{EF}}_{i,j}={\mathit{EF}}_{i,j}(direct)+{\mathit{EF}}_{i,j}(indirect),$$ [2]

where $E{F}_i$ represents the environmental footprint indicator per unit mass of the product where i refers to certain type of footprint like WF (in m³/t) or CF (in t CO₂ eq/t); $E{F}_{i,j}$ is the $E{F}_i$ of the life-cycle stage j, t CO₂ eq/t or m³/t; $P$ denotes the annual corrugated box packing consumption related to the express delivery industry, t (SI Appendix, Fig. S1); $E{F}_{i,j}(direct)$ and $E{F}_{i,j}(indirect)$ indicate the unit direct and indirect environmental footprint i, of the life-cycle stage j, respectively, t CO₂ eq/t or m³/t. Life-cycle stage j corresponds to the RMS, BMS, USS, and ELS, respectively.

RMS.

The environmental footprints accounting for the RMS consists of both virgin pulp and recycled pulp production processes. Depending on the source of raw materials, virgin pulp can be classified into three types: domestic wood pulp, imported wood pulp from 11 different countries, and domestic straw pulp, and the recycled pulp is divided into domestic recycled pulp and imported recycled pulp from 15 different countries (imported recycled pulp refers to waste paper imported from abroad, while the pulping process is performed in the domestic factories). The environmental footprints accounting of recycled pulp includes pulping process and environmental debit allocated from virgin pulp based on the fiber mass flow (49). The environmental footprints of the virgin material stage can be calculated as follows:

$${\mathit{EF}}_{i,RMS}=E{F}_{i,v}\times A_u\times C+{\mathit{EF}}_{i,v}\times (1-C)+C\times E{F}_{i,r},$$ [3]

where $E{F}_{i,RMS}$ is the unit environmental footprints of corrugated boxes in the RMS, t CO₂eq/t or m³/t; $E{F}_{i,v}$ is the unit environmental footprint i of virgin pulp production from natural resource, t CO₂eq/t or m³/t; $C$ is the recycled pulp content of corrugated boxes; and $A_u$ is the environmental debit allocation factor related to the fiber cycle numbers throughout the paper production system. Using the one-parameter model proposed by Meinl et al. (53), the number of fibre cycles in the whole paper production system can be estimated. Therefore, $A_u$ can be expressed as a function related to the waste paper utilization rate $U$:

$$A_u=\frac{\mathit{MNU}}{MFA+MNU-1}=\frac{1}{1+U},$$ [4]

where $MNU$ represents the average number of times that the fibres in the paper production system were used by the ELS; MFA represents the number of past cycles of the fibres; $U$ is the waste paper utilization rate, which was defined as utilization of paper for recycling plus the net trade of paper for recycling compared to paper consumption.

The unit environmental footprints of virgin pulp are calculated by weighting the unit environmental footprints of three different types of virgin pulp (domestic wood pulp, domestic straw pulp, and imported wood pulp) by the proportion of pulp consumption per year. The pulp yield rate, resource consumption, pollutant emissions, and chemical inputs of different virgin pulps vary depending on the fibre content of different raw materials (straw and wood) and the pulping process used. In this study, we calculated the unit environmental footprints of wood pulp and three kinds of straw pulp produced in China, as well as wood pulp from 11 countries which accounted for more than 94% of total wood pulp imports of China during 2007 to 2021 (SI Appendix, Fig. S3). The accounting process for each pulp includes the environmental footprint from raw material production (wood and crop straw production), pulping process, energy input, chemical input, and transportation. Calculations for the unit environmental footprints of virgin pulp $E{F}_{i,v}$ and recycled pulp $E{F}_{i,r}$ are expressed by Eqs. 5 and 6:

$${\mathit{EF}}_{i,v}=\sum _{v}pa\times {\omega }_v\times \left(\frac{{\mathit{EF}}_{i,v,m}+{\mathit{EF}}_{i,v,t}}{{\mathit{pu}}_v}+{\mathit{EF}}_{i,v,p}+E{F}_{i,v,e}+{\mathit{EF}}_{i,v,c}\right),$$ [5]

$${\mathit{EF}}_{i,r}=\sum _{r}pa\times {\omega }_r\times \left(\frac{{\mathit{EF}}_{i,r,t}}{{\mathit{pu}}_r}+{\mathit{EF}}_{i,r,p}+E{F}_{i,r,e}+{\mathit{EF}}_{i,r,c}\right),$$ [6]

where $pa$ is unit pulp demand for corrugated box production, t/t (SI Appendix, Table S1); ${\omega }_v$ represents the proportion of virgin pulp, k, consumed in the total virgin pulp by China’s pulp and paper industry; ${\omega }_r$ represents the proportion of recycled pulp r consumed in the total recycled pulp by China’s pulp and paper industry; $E{F}_{i,v,m}$ represents the unit environmental footprint i of raw material production of virgin pulp v, t CO₂eq/t or m³/t; $E{F}_{i,v,t}$ and $E{F}_{i,r,t}$ represent the unit environmental footprint i of virgin raw material k and recycled raw material r collection and transportation, respectively, t CO₂eq/t or m³/t; $p{u}_v$ and $p{u}_r$ represent the raw material pulp yield of virgin pulp v and recycled pulp r, respectively, that is, the consumption of the raw material (virgin or recycled) to produce 1 t pulp, t/t; $E{F}_{i,v,p}$ and $E{F}_{i,r,p}$ represent the unit environmental footprint i of virgin pulp v and recycled pulp r pulping process, respectively, t CO₂eq/t or m³/t; $E{F}_{i,v,e}$ and $E{F}_{i,r,e}$ represent the environmental footprint i of energy input in virgin pulp v and recycled pulp r, respectively, t CO₂eq/t or m³/t; $E{F}_{i,v,c}$ and $E{F}_{i,r,c}$ refer to the unit environmental footprint i of chemicals input in virgin pulp v and recycled pulp r, respectively, t CO₂eq/t or m³/t; v refers to different types of virgin pulp, including domestic wood pulp, domestic straw pulp, and imported wood pulp; r refers to two kinds of recycled pulp, which is domestic recycled pulp, imported recycled pulp.

BMS.

The BMS is divided into three sub-processes, including boxboard, corrugated medium (CM) production, and the assembling process of two kinds of paperboard. The unit environmental footprints of the BMS, $E{F}_{i,BMS}$, can be expressed as follows:

$${\mathit{EF}}_{i,BMS}=\sum _k(E{F}_{i,k,p}+E{F}_{i,k,e}+{\mathit{EF}}_{i,k,c}+{\mathit{EF}}_{i,k,t}),$$ [7]

where $E{F}_{i,k,p}$ represents the unit environmental footprint i of sub-process k; $E{F}_{i,k,e}$ represents the environmental footprint i of energy input in sub-process k; $E{F}_{i,k,c}$ represents the environmental footprint i of chemicals input in sub-process k; $E{F}_{i,k,t}$ represents the unit environmental footprint i of sub-process k transportation; k refers to sub-processes in the BMS, including boxboard making, CM making and assembling process.

USS.

The USS stage involved two parts: inter-city and intra-city express delivery. The unit environmental footprints in this stage are dominated by the transportation process for the express delivery services (39, 54), which can be expressed by

$${\mathit{EF}}_{i,USS}=\sum _l{\omega }_l\times {\mathit{EF}}_{i,l}\times d_l,$$ [8]

where $E{F}_{i,USS}$ represents the unit environmental footprint i in the USS. $E{F}_{i,l}$ represents the average unit environmental footprint i of express delivery service l transportation. l indicates two types of parcel delivery modes. $d_l$ is the transportation distance of express delivery service l. The provincial express delivery transportation mode and distance for each year were calculated based on a spatially based dynamic model (39).

ELS.

The ELS includes three waste disposal options: recycling, incineration, and landfill. The corresponding unit environmental footprints, $E{F}_{i,ELS}$, can be expressed as follows:

$${\mathit{EF}}_{i,ELS}=\left(1-R\right)\times \sum _d{w}_d\times {\mathit{EF}}_d-R\times A_r\times {\mathit{EF}}_{i,v},$$ [9]

where $R$ is the recycling rate of corrugated boxes used by the express delivery industry of China; $w_d$ represents the proportion of waste disposal option d; $E{F}_d$ represents the unit environmental footprints of waste disposal option d; and $A_r$ is the recycling environmental credit allocation factor, which can be calculated using Eq. 10.

$$A_r=\frac{MNU-1}{MFA+MNU-1}=\frac{U}{1+U}.$$ [10]

Environmental footprint accounting method for each stage and all data used are detailed in SI Appendix.

Driving Factor Analysis.

The Logarithmic Mean Divisia Index (LMDI) method (54, 55) was used to analyze the drivers of environmental footprint changes of corrugated boxes in China’s express industry. The LMDI method allows for optimal decomposition, i.e., no unexplained residual terms, consistent aggregation, and satisfactory additivity, and is widely used in studies related to carbon emissions and water consumption (56–58).

The environmental footprints of the corrugated boxes life cycle are divided into four stages, which can be expressed as follows:

$${\mathit{EF}}_i=\sum _j\sum _m{\mathit{EF}}_{i,j,m},$$ [11]

$${\mathit{EF}}_{i,1}=\sum _m\frac{{\mathit{EF}}_{i,1,m}}{P_{1,m}}\times \frac{P_{1,m}}{P_1}\times P_1=\sum _{m}R(EF{I}_{i,m})\times R(S_m)\times R(P_1),$$ [12]

$${\mathit{EF}}_{i,2}=\sum _m\frac{{\mathit{EF}}_{i,2,m}}{P}\times P=\sum _{m}B(EF{I}_{i,m})\times B(P),$$ [13]

$${\mathit{EF}}_{i,3}=\sum _m\frac{{\mathit{EF}}_{i,3,m}}{P_{3,m}}\times \frac{P_{3,m}}{P}\times P=\sum _{m}U(EF{I}_{i,m})\times U(S_m)\times U(P),$$ [14]

$${\mathit{EF}}_{i,4}=\sum _m\frac{{\mathit{EF}}_{i,4,m}}{P_{4,m}}\times \frac{P_{4,m}}{P}\times P=\sum _{m}E(EF{I}_{i,m})\times E(S_m)\times E(P),$$ [15]

where $E{F}_i$ is the life-cycle environmental footprint of corrugated boxes used in China’s express delivery industry, t CO₂eq or m³; i represents different environmental footprint indicators, including CF, blue WF, green WF, and gray WF. Life-cycle stage j corresponds to the RMS, BMS, and ELS, respectively. $E{F}_{i,1,m}$ is the environmental footprint i of pulp m in the RMS, where m = 1 to 5 represents different types of pulp (including three kinds of virgin pulp and two kinds of recycled pulp); $P_{1,m}$ is the consumption of pulp m in RMS; $P_1$ is the total pulp consumption in RMS. $E{F}_{i,2,m}$ is the environmental footprint i of sub-process m in BMS, where m = 1 to 3 represents three sub-processes including box board production, CM production and assembling in BMS; $P$ is the corrugated box packing consumption related to express delivery industry, t; $E{F}_{i,3,m}$ is the environmental footprint i of express delivery m in the USS, where m = 1 to 2 represents two types of express delivery service (inter-city and intra-city) in the USS. ${\mathit{EF}}_{i,4,m}$ is the environmental footprint i of waste disposal m in the ESS, where m = 1 to 3 represents three waste disposal options (landfill, incineration, and recycling) in the ESS. According to Eq. 12, environmental footprints in RMS are broken down into three factors: the unit environmental footprints of different types of pulp $R\left(EF{I}_{i,m}\right)=E{F}_{i,1,m}/P_{1,m}$ to measure the environmental footprints intensity effects in the RMS, the proportion of pulp m consumption in the total pulp consumption $R(S_m)=P_{1,m}/P_1$ to represent the raw material structure effects in the RMS, total pulp consumption $R(P_1)=P_1$, representing the scale effect in the raw material. According to Eq. 13, environmental footprints in the BMS are divided into two categories: the unit environmental footprints of different sub-processes $B(EF{I}_{i,m})=E{F}_{i,2,m}/P$ to measure the environmental footprints intensity effects in the BMS, and the corrugated box packing consumption $P=B(P)$ represents the scale effects in the BMS. According to Eq. 14, environmental footprints in the USS are divided into three parts: the unit environmental footprints of two different types of delivery service $U(EF{I}_{i,m})=E{F}_{i,3,m}/P$ to measure the environmental footprints intensity effects in the USS, and the proportion of express delivery service mode m in the total express delivery volume $U(S_m)=P_{3,m}/P$  to represent the express delivery mode structure effects in the USS, total express delivery volume $U(P)=P,$ representing the scale effect in the USS. According to Eq. 15, environmental footprints in ELS are divided into three categories: the unit environmental footprints of different waste disposal options $E(EF{I}_{i,m})=E{F}_{i,4,m}/P_{4,m}$ to measure the environmental footprint intensity effects in the ELS, the proportion of waste disposal option m, $E(S_m)=P_{4,m}/P$ representing the waste disposal structure effects in the ELS, and the corrugated box packing consumption $P=E(P)$ represents the scale effects in the ELS.

Using the LMDI decomposition, the change of environmental footprints in three life-cycle stages during year t-1 to year t is calculated as follows:

$$\mathrm{\Delta }{\mathit{EF}}_i={\mathit{EF}}_i^t-{\mathit{EF}}_i^{t-1}=\sum _j\mathrm{\Delta }{\mathit{EF}}_{i,j},$$ [16]

$$\begin{array}{c}\mathrm{\Delta }{\mathit{EF}}_{i,1}=\sum _m{w}_{i,1,m}ln\left(\frac{R{\left({\mathit{EFI}}_{i,m}\right)}^t}{R{\left({\mathit{EFI}}_{i,m}\right)}^{t-1}}\right)\hfill \\ +{\displaystyle \sum _m}{w}_{i,1,m}ln\left(\frac{R{\left(S_m\right)}^t}{R{\left(S_m\right)}^{t-1}}\right)+{\displaystyle \sum _m}{w}_{i,1,m}ln\left(\frac{R{\left(P_1\right)}^t}{R{\left(P_1\right)}^{t-1}}\right)\hfill \\ = \mathrm{\Delta }{\mathit{EF}}_{R\left({\mathit{EFI}}_i\right)}+\mathrm{\Delta }{\mathit{EF}}_{R\left(S_i\right)}+\mathrm{\Delta }{\mathit{EF}}_{R\left(P_i\right)},\hfill \end{array}$$ [17]

$$\begin{array}{c}\mathrm{\Delta }{\mathit{EF}}_{i,2}={\displaystyle \sum _m}{w}_{i,2,m}ln\left(\frac{B{\left({\mathit{EFI}}_{i,m}\right)}^t}{B{\left({\mathit{EFI}}_{i,m}\right)}^{t-1}}\right)+{\displaystyle \sum _m}{w}_{i,2,m}ln\left(\frac{P^t}{P^{t-1}}\right)\hfill \\ = \mathrm{\Delta }{\mathit{EF}}_{B({\mathit{EFI}}_i)}+\mathrm{\Delta }{\mathit{EF}}_{B(P_i)},\hfill \end{array}$$ [18]

$$\begin{array}{c}\mathrm{\Delta }{\mathit{EF}}_{i,3}={\displaystyle \sum _m}{w}_{i,3,m}ln\left(\frac{U{\left({\mathit{EFI}}_{i,m}\right)}^t}{U{\left({\mathit{EFI}}_{i,m}\right)}^{t-1}}\right)\hfill \\ +{\displaystyle \sum _m}{w}_{i,3,m}ln\left(\frac{U(S_m{)}^t}{U(S_m{)}^{t-1}}\right)+{\displaystyle \sum _m}{w}_{i,3,m}ln\left(\frac{U{\left(P\right)}^t}{U{\left(P\right)}^{t-1}}\right)\hfill \\ = \mathrm{\Delta }{\mathit{EF}}_{U({\mathit{EFI}}_i)}+\mathrm{\Delta }{\mathit{EF}}_{U(S_i)}+\mathrm{\Delta }{\mathit{EF}}_{U(P_i)},\hfill \end{array}$$ [19]

$$\begin{array}{c}\mathrm{\Delta }{\mathit{EF}}_{i,4}={\displaystyle \sum _m}{w}_{i,4,m}ln\left(\frac{E{\left({\mathit{EFI}}_{i,m}\right)}^t}{E{\left({\mathit{EFI}}_{i,m}\right)}^{t-1}}\right)\hfill \\ +{\displaystyle \sum _m}{w}_{i,4,m}ln\left(\frac{E{\left(S_m\right)}^t}{E{\left(S_m\right)}^{t-1}}\right)+{\displaystyle \sum _m}{w}_{i,4,m}ln\left(\frac{E{\left(P\right)}^t}{E{\left(P\right)}^{t-1}}\right)\hfill \\ = \mathrm{\Delta }{\mathit{EF}}_{E\left({\mathit{EFI}}_i\right)}+\mathrm{\Delta }{\mathit{EF}}_{E\left(S_i\right)}+\mathrm{\Delta }{\mathit{EF}}_{E\left(P_i\right)},\hfill \end{array}$$ [20]

where $w_{i,j,m}=(E{F}_{i,j,m}^t-E{F}_{i,j,m}^{t-1})/ln(E{F}_{i,j,m}^t/E{F}_{i,j,m}^{t-1})$ is a weighting factor called the logarithmic mean weight. $\mathrm{\Delta }E{F}_{R(EF{I}_i)}$, $\mathrm{\Delta }E{F}_{R(S_i)}$, and $\mathrm{\Delta }E{F}_{R(P_i)}$ represent the environmental footprint i change in the RMS corresponding to change in the unit environmental footprint of different raw material, shift in raw material consumption structure, and the change in consumption of raw material, respectively. $\mathrm{\Delta }E{F}_{B(EF{I}_i)}$ and $\mathrm{\Delta }E{F}_{B(P_i)}$ are the environmental footprint i changes in the BMS corresponding to change in the unit environmental footprint of different sub-processes and change in consumption of corrugated boxes, respectively. $\mathrm{\Delta }E{F}_{U(EF{I}_i)}$, $\mathrm{\Delta }E{F}_{U(S_i)}$, and $\mathrm{\Delta }E{F}_{U(P_i)}$ represent the environmental footprint i change in the USS corresponding to change in the unit environmental footprint of different express delivery service mode, shift in express delivery service mode structure, and the change in express delivery volume, respectively. $\mathrm{\Delta }E{F}_{E\left(EF{I}_i\right)}$, $\mathrm{\Delta }E{F}_{E\left(S_i\right)}$, and $\mathrm{\Delta }E{F}_{E\left(P_i\right)}$ refer to the environmental footprint i change in the ELS corresponding to change in the unit environmental footprint of different waste disposal options, change in the structure of waste treatments, and change in consumption of corrugated boxes, respectively.

Scenario Analysis.

To further explore the changing trends of the environmental footprints and the potential for resource conservation and emission reduction in the life cycle of corrugated boxes packaging in China’s express delivery industry, we combined the historical trends in the structure and key factors of the previous driving force analysis, selected raw material restructuring and energy restructuring at the production end, single-piece corrugated box use and optimizing logistics process at the consumption end, and waste paper recycling at the waste treatment stage in the post-consume end, respectively. Different future scenarios are set and the future environmental footprint change pattern is predicted and analysed. The future volume trend of express deliveries is described by a logistic growth model (39). China’s express business volume reaches its peak in 2035 (SI Appendix, Fig. S20); therefore, the forecast time period is selected to be 2022 to 2035, and the specific express volume forecast data are shown in SI Appendix. SI Appendix, Table S14 shows the contents and basis of different scenarios.

Sensitivity Analysis.

Allocation procedure is a key issue in life-cycle environmental footprint accounting, especially for the situation to solving multi-functionality problem in the systems with open-loop recycling (59). Here, we conducted three different allocation method covered by ISO14067__reffibre (49), the Production Environmental Footprint (EC 2013) (59) and GHG Protocol or PAS 2050 in the life-cycle environmental footprints accounting of corrugated boxes for express delivery services in China to discuss the impacts of different allocation procedures on the results. Whether or how to consider the propagation of environmental burden associated with recycled materials between multiple life cycles is the main difference of those three methods. The allocation procedures presented in the GHG Protocol and PAS 2050 accounts neither for the credits for recycling nor the debits for the use of recycled material (a “cut-off” approach), which was shown in Eq. 21. As shown in Eqs. 21 and 22, the allocation method from PEF2018 and ISO 14067__reffibre applied fixed and dynamic factors to share the environmental impacts of recycling credits and debits in different life cycles, respectively. We adopted all those three methods in our studies, and then estimated the weighted difference ($E_{diff}$) between these results ($E_i$) and the results of the original method ($E_0$) used in this study (ISO 14067__reffibre), as shown in Eq. 24, to obtain the sensitivity of different allocation procedure to the results (SI Appendix, Table S16).

$$\begin{array}{c}{\mathit{EF}}_i = {\mathit{EF}}_{i,v} \times \left(1-C\right) + C \times E{F}_{i,r} + {\mathit{EF}}_{i,BS}\hfill \\ + {\mathit{EF}}_{i,US} + \left(1-R\right) \times {\displaystyle \sum _d}{w}_d \times {\mathit{EF}}_d,\hfill \end{array}$$ [21]

$$\begin{array}{c}{\mathit{EF}}_i = E{F}_{i,v} \times A \times C + {\mathit{EF}}_{i,v} \times \left(1-C\right)+C \times E{F}_{i,r} + {\mathit{EF}}_{i,BS} + {\mathit{EF}}_{i,US}\hfill \\ + \left(1-R\right) \times {\displaystyle \sum _d}{w}_d \times {\mathit{EF}}_d-R \times A \times {\mathit{EF}}_{i,v},\hfill \end{array}$$ [22]

$$\begin{array}{c}{\mathit{EF}}_i = E{F}_{i,v} \times A_u \times C+{\mathit{EF}}_{i,v} \times \left(1-C\right) + C \times E{F}_{i,r} + {\mathit{EF}}_{i,BS}\hfill \\ + {\mathit{EF}}_{i,US} + \left(1-R\right) \times {\displaystyle \sum _d}{w}_d \times {\mathit{\text{EF}}}_d-\text{ R } \times A_r \times {\mathit{\text{EF}}}_{i,v},\hfill \end{array}$$ [23]

$$E_{\mathit{\text{diff}}}=\frac{E_0-E_i}{E_i},$$ [24]

where $E{F}_i$ is the unit environmental footprints i of corrugated boxes, t CO₂eq/t or m³/t; $E{F}_{i,v}$ is the unit environmental footprint i of virgin pulp production from natural resource, t CO₂eq/t or m³/t; $C$ is the recycled pulp content of corrugated boxes; and $A_u$ is the environmental debit allocation factor related to the fibre cycle numbers throughout the paper production system. $E{F}_{i,r}$ is the unit environmental footprints of recycled pulp; $R$ is the recycling rate of corrugated boxes used by the express delivery industry of China; $w_d$ represents the proportion of waste disposal option d; $E{F}_d$ represents the unit environmental footprints of waste disposal option d; and $A_r$ is the recycling environmental credit allocation factor, which can be calculated using Eq. 10. A is a fixed factor.

Uncertainty Analysis.

Uncertainties in the current environmental footprints accounting are mainly resulted from consumptions about the use of different size corrugated boxes in express delivery industry, the historical development of corrugated box production efficiencies, and economic data on national and provincial raw materials and intermediate products trade along the entire supply chain. To assess the uncertainties, we choose 11 key parameters (SI Appendix, Table S15) with high fluctuation. The parameters include the energy consumption, the water withdraw, the amount of chemical oxygen demand (COD) emissions before sewage treatment, and the COD removal rate for per unit key intermediate products in production supply chain stage (including RMS and BMS), the use fraction of corrugated boxes in the parcel packaging and the share of four types corrugated boxes in the use stage, and the corrugated boxes recycling rate in the ELS. The resulting uncertainties of those parameters in year 2015 were then applied in Monte-Carlo simulation to quantify the uncertainties of the environmental footprints accounting. The sensitivities of the results to each of these parameters are also evaluated (SI Appendix, Figs. S28 and S29).

The results from 5,000 Monte-Carlo runs indicate that the unit environmental footprints (CF, gray WF, blue WF, green WF) follow normal distributions lying between 2.7 and 3.3 t CO₂eq/t, 289 and 673 m³, 22 and 33 m³, and 221 and 269 m³, respectively (95% level of confidence). The corresponding total environmental footprints also follow normal distributions with a mean (μ) of 9.4 Mt CO₂eq (σ = 0.9), a mean (μ) of 1.4 Gm³ (σ = 0.3), a mean (μ) of 84 Mm³ (σ = 10), and a mean (μ) of 774 Mm³ (σ = 73), respectively. The coefficient of variations of both the unit and total CF, blue WF, gray WF of corrugated boxes range between 5.1% and 11.9%, which reveals the robustness of the results. The gray WFs have relative high coefficient of variations with 22.4% for the unit and 23.8% for the total that are devoted to the highly uncertain of cleaner production level between large-scale and small-scale pulp and paper enterprise in China. The national scale-specific data for all enterprise are not available, which also highlights the challenges of conducting a comprehensive and transparent environmental footprint assessment along entire supply chain.

Supplementary Material

Appendix 01 (PDF)

Data, Materials, and Software Availability

The average weight of corrugated boxes in a single express delivery was compiled from field research data reported by Duan et al. (47) and Su et al. (25). Express delivery volume data were obtained from publicly available data from State Post Bureau of China. Process technology flow data for different domestic virgin pulp and paperboard were obtained from the China Pulp and Paper Industry Clean Production Standard, and the 12th and 13th Five-Year Plans for the pulp and paper industry. Data on corrugated box assembling process were obtained from the Ecoinvent v3.6 database (60). Data on the energy structure of corrugated box production were obtained from China Energy Yearbook, those on fiber raw material structure and waste paper utilization and recycling rate of pulp and paper industry from China Paper Industry Yearbook, and those on import and export volume of wood pulp and waste paper from UN trade data. The distance of ocean freighter transport was obtained from the dataset SEA DISTANCES/PORT DISTANCES (https://sea-distances.org) (61). Detailed information on data sources can be obtained from the SI Appendix. All study data are included in the article and/or SI Appendix.