Responses to the comments of Referee #1

Comment: This manuscript investigated the anthropogenic vanadium emission inventory in China during 2015-2019 to various environmental receptors including atmosphere, soil and water. Considering its integrated contents and the scope of Atmospheric Chemistry and Physics, it might be more suitable for an environmental journal rather than ACP.

Reply: We would like to express our sincere gratitude to the referee for raising this question. The atmosphere pollution is indeed the major focus of our inventory development for vanadium emission. As shown in the result, vanadium emission contributed predominantly to atmospheric pollution (97.3%). However, the emission inventories need to capture the full aspects of each emission source, including direct discharge to other environmental receptors. In addition, vanadium flux released into atmosphere may interact with other environmental media. Therefore, focusing only atmospheric emission may overlook the significance of interconnectedness of different environmental media. With in depth understanding of vanadium emission across different medias, and the patterns of each emission source, more effective and targeted measures will be proposed by policy makers. Lastly, we noticed there are articles (doi.org/10.5194/acp-16-4451-2016; doi.org/10.5194/acp-20-16117-2020) published in ACP, which covered the emission flux directed to environmental media other than atmosphere. Therefore, we believe that our work is relevant to the scope of ACP journal.

Some comments are suggested for considering:

The writing like statements should be clarified overall. For instance, Abstract Line 12-13: Is the flux value for the period of five years or a year?

Reply: Thank you for pointing this out, we have clarified it. The flux values were resulted from five years period (2015-2019).

Line 12-13: Cumulative emission flux modelling has shown that 211094 t, 3725 t, and 0.1 t of vanadium were discharged into atmosphere, soil and water during 2015-2019, respectively.

To avoid the limitation of a local investigation, the scientific implications should be strengthened. Such as Abstract Line 14-15, what’s the international or scientific implications for these provinces with largest emissions?

Reply: All those provinces with largest emissions were densely populated and highly industrialized in China. In addition, the combined emission of vanadium in those provinces also contributed greatly to both national and global vanadium emission inventories. We have incorporated the following statement to strengthen the scientific implication of local results.

Line 14-18: The spatial and temporal of vanadium emission is strongly governed by industrial activity levels across regions. Shandong, Liaoning, Hebei, Guangdong and Hunan, with dense population and great level of industrial activities, are among the largest provincial emitters, contributing 35.8% of national vanadium inventories. Provinces in North China and Northwestern China increase their emission contribution pertinent to raw coal combustion by 719 t and 316 t, respectively, due to their high energy demand.

Line 135-142: In comparison, according to the most up to date research on global vanadium emission (Schlesinger et al., 2017), vanadium released from both stationary and mobile source of oil combustion jointly constituted over 60% of vanadium emission inventory, while coal burning constituted only 26% of total output. To explain the discrepancy, the development of global inventory employed mainly the available data of industrial activities in the United States. Unlike the United States, coal combustion played pivotal roles in China’s energy sector, and vanadium output pertinent to oil extraction was much less. Moreover, vanadium releasing fraction, emission reducing technology, and vanadium content in raw materials varied significantly across regions. To improve the inventory, it is essential to perform local level investigation in different regions.

Line 198-202: In general, a substantial amount of vanadium emission was found in northern, eastern coastal, southern and south western provinces. Shandong, Liaoning, Hebei, Guangdong, and Hunan, the most industrially developed provinces in China, were among the largest provincial emitters with cumulative vanadium discharge over 10000 t (Fig. S3a), accounting for over 35.8% of total national emission in combined.

Line 249-253: The development of vanadium emission inventory at provincial level will serve as key basis to policy making, which would warrant more targeted efforts on major provincial emitters for different emission source. The detailed results on vanadium emission will also provide benchmark for comparing the effectiveness of administrative and technological measures for pollution control. In addition, local results will contribute greatly to the development of global emission inventory, which requires in depth understanding of sources and magnitudes of emission.

If submitted to ACP, this paper should focus on emissions to atmosphere in detail with depth.

Reply: We very much appreciate your comment. The inventory indeed was developed with aim to illustrate the vanadium emission to multiple environment medias, with primary focus on atmospheric emission. Firstly, the inventory provided a quantitative assessment of vanadium emission, including coal combustion, stationary oil burning, transportation, industrial production (steelmaking and glass production), and waste handling (MSW incineration), consisted predominantly vanadium flux to atmosphere (97.3%), as indicated by Figure 1.

Line 114-115: During 2015-2019, 98.3% vanadium generated from anthropogenic processes was directly discharged into the atmosphere, leading to a total emitted vanadium content of 211094 t.

The paper also delineated the temporal evolvement of vanadium emission from major emission sources, highlighting that the major emission contributors are coal combustion and oil burning in combined, both of which were entirely responsibly for atmospheric discharge. In addition, we have drawn parallel with atmospheric vanadium emission on global scale, providing more insight by comparing the difference in contribution fraction by emission sources.

Line 136-143: Investigation on global vanadium emission also accounted the fossil fuel burning as the primary source of atmospheric vanadium discharge (Schlesinger et al., 2017), with stationary and mobile source of oil combustion jointly constituting over 60% of vanadium emission flux, while coal burning made up of only 26% of total output. To explain the discrepancy, the development of global inventory employed mainly the available data of industrial activities in the United States. However, coal combustion played pivotal roles in China’s energy sector, and vanadium output pertinent to oil extraction was much less. The contrast could also be attributed to variation in vanadium releasing fraction, emission reducing technology, and vanadium content in raw materials in different regions. To improve the inventory, it is essential to perform local level investigation in different regions.

For each emission source contributing to atmosphere, and other medias, we have also presented their relative contributions to the emission inventory.

Line 145-149: The relative contribution of coal combustion to the emission inventory slightly decreased by 2.0%.

Line 150: corresponding to a decrease in relative contribution by 8.5%.

Line 155: which led to an increase in relative contribution by 9.7%, 6.4% and 28%, respectively.

Line 165: along with apparent increase in relative contribution to the inventory by 30.6%, -2.0%, and 37.9%.

Furthermore, we have enhanced our discussion on the implications of atmospheric emission results, in conjunction with the impact of the "Clean Air Act". We have explored the roles this policy has played in vanadium emissions and the limitations of the current study, offering insights into future improvements in data acquisition.

Line 146-148: The emission trend of coal combustion was similar to observation in contemporary study, where atmospheric vanadium emission significantly decreased compared to pre-2013 periods as result of introduction of “Clean Air Act”, followed by a relatively constant trend (Bai et al., 2021).

Line 168-176: However, the trend of sulfur content remained flattened in transportation derived emission (Zheng et al., 2018), which implied an increased utilization of vanadium-based catalyzer for exhaust gas treatment (National Bureau of Statistics of China, 2020). Despite the overall increase in industrial activity levels, according to a comprehensive study on anthropogenic emissions trend after the introduction of “Clean Air Act”, energy and industrial manufacturing sectors underwent drastic reduction in pollutant emission (e.g., SO2, NOx) (Zheng et al., 2018), while emission derived from transportation (e.g., NOx, CO, NMVOC) was maintained relatively steady. The result suggested that potential impact of improvement in technological and administrative measures on pollutant emission may counterbalance or even exceed the impact induced by increased activity levels, especially for combustion of coal and petroleum products affiliated with energy and industrial sectors. In comparison, change in activity level more likely affected the trend of transportation derived emission.

Line 292-295: Consequently, the industrial activity level appears to be the primary influencing factor of vanadium emission levels, overshadowing the potential impact of variations in other parameters, such as removal efficiency of pollution control technologies and vanadium content in raw materials, which also varied across regions and processes. To further refine the inventory, acquiring region-specific data and conducting onsite investigations will be essential in our future efforts.

Is year 2015-2019 representative for interannual variation of temporal analysis? Why not longer periods?

Reply: The year of 2015-2019 period is representative considering the following reasons: (1) This time period provided representation in emission trend as result of “Clean Air Act” introduced since the end of 2013. Under the new act, major sectors pertinent to vanadium emission undertook effort to reduce emissions, including power plant, steelmaking, petrochemical, steel and glass production (doi.org/10.5194/acp-18-14095-2018). (2) 2020-2022 marked a period of trough in all industrial activities due to the pandemic, therefore we treated these years as emission anomaly, which is not representative. We have included the following sentences to give more background regarding the study period.

Line 50-52: Moreover, China have adopted “Clean Air Act” since 2014, with aim to reduce the atmospheric emission in wide ranges of industrial activities (Zheng et al., 2018). However, the impact of implementing this act on vanadium emission trend was still underexplored.

Line 62-63: The goal of this study is to establish the bottom-up inventories of vanadium emission from anthropogenic activities, targeting specifically the period from 2015 to 2019 in China post to the introduction of “Clean Air Act”.

Line 395: Zheng, B., Tong, D., Li, M., Liu, F., Hong, C., Geng, G., Li, H., Li, X., Peng, L., Qi, J., Yan, L., Zhang, Y., Zhao, H., Zheng, Y., He, B., Zhang, Q.: Trends in China’s anthropogenic emissions since 2010 as the consequence of clean air actions, Atmos. Chem. Phys., 18, 14095-14111, https://doi.org/10.5194/acp-18-14095-2018, 2018.

Is seasonal change more meaningful in this study?

Reply: Unfortunately, the seasonal variation data is very limited as we performed our study specifically based on yearbook of industrial activities. However, for future effort, we can explore the seasonal spatiotemporal variation in selected regions where seasonal data is available.

Line 295-299: In this study, the spatiotemporal variation pattern of vanadium emission was characterized based on yearly recorded datasets. However, there is still room to improve the accuracy of emission inventory. Seasonal investigation may help in identifying the source specific variation due to altering energy demand (e.g., thermal power plant) or weather influence (e.g., precipitation, strong wind) following seasonal patterns.

Were their spatial and temporal differences in emission sources? How to consider them in the national emission inventory calculation?

Reply: Yes, industrial activity levels pertinent to different emission sources varied in spatial and temporal manner. In the formula of calculating the emission inventories, emission related parameters such as treatment efficiency, vanadium release fraction, and dynamic emission factors, were all obtained from previous literature, and applied into the equation in average values. Therefore, even though our results suggested that the variation in industrial activity level strongly affected the change in vanadium emission, their actual impact may be offset by difference in technological impact (e.g., one region employed better pollution abatement technology) or using cleaner fuel.  We have incorporated the following sentence for explanation.

Line 173-176: The result suggested that potential impact of improvement in technological and administrative measures on pollutant emission may counterbalance or even exceed the impact induced by increased activity levels, especially for combustion of coal and petroleum products affiliated with energy and industrial sectors. In comparison, change in activity level more likely affected the trend of transportation derived emission.

Line 298-303: Furthermore, we have utilized average values of reported emission related parameters. Consequently, the industrial activity level appears to be the primary influencing factor of vanadium emission levels, overshadowing the potential impact of variations in other parameters, such as removal efficiency of pollution control technologies and vanadium content in raw materials, which also varied across regions and processes. To further refine the inventory, acquiring region-specific data and conducting onsite investigations will be essential in our future efforts.

Responses to the RC2

Comment: The manuscript presents and discusses, to some extent, the calculated anthropogenic Vanadium emission inventory at China for the period 2015-2019. The emissions are reported in terms of compartments, atmosphere, water and soil over the years and for the 31 Chinese provinces/7 regions. The content is of great relevance considering China as one of the main, perhaps the largest economy in the planet with strong and active industry and energy production sectors. However, in my opinion, the paper lacks in discussing their findings. For example, the authors did not put the findings into a world perspective, did not compare with other studies, etc., which could enhance the importance of such results. Also, the methodology is poorly described. In my opinion, the paper requires substantial improvement in order to be accepted.

Reply: We appreciate the reviewer’s constructive feedback. We also acknowledge the importance of providing a more comprehensive discussion of our findings by comparing them with global studies on vanadium emissions. To address this concern, we have significantly enhanced the discussion section of our manuscript, with emphasis on the following three perspectives:

• Comparison with global emission studies: Including a detailed comparison of our findings with global vanadium emission studies, highlighting regional differences in emission sources.
• More insights into the regional contexts within China: Expanding discussions on the variation in emission patterns across Chinese regions/provinces, emphasizing the impact of environmental regulations, emission control technologies, and socioeconomic conditions.
• Future improvements: evaluate the uncertainties of emission inventory, identify the source of uncertainties, and propose improvement in method of data collection and handling to increase the inventory’s credibility.

These revisions enrich our manuscript by situating our findings in a broader context and addressing the reviewer’s concerns. We hope these improvements meet the expectations and enhance the overall quality of our work. The detailed revisions are as follow:

General points

1) In the methodology section, it is important that the authors provide far more details on how the inventory was built. For example, what were the data sources for each category presented in the Supplements? The inventory described in the supplemental material, is it available somewhere, is it public? Please include some information on how those values reported on the tables were obtained, were they measured, were they modelled? On page 10 and line 233-236 the authors mention about the data availability, please provide more description in the methodology section about that. Some explanation on why this specific period was selected would be interesting.

Reply: Thank you for the insightful comments. We have explicitly provided all references for activity levels related to vanadium emission sources in the Supplement and Appendix, alongside with specific citation in the manuscript texts. The activity levels were retrieved primarily from publicly available sources, such as statistical yearbook of China’s industrial activities, including coal industry, energy, automobile, light industry, steelmaking, and urban-rural construction, etc. We have also illustrated that how the inventory data were calculated in a succinct manner. In addition, we have explicitly described the potential limitation of inventory development due to limited data availability, and what we do to address this. Lastly, we have briefly explained the reason to select this time period for our study.

Line 80-88: For each emission source, activity data at provincial level were retrieved from yearbooks of China’s industrial activities, including coal (Appendix SS-A), energy (Appendix SS-B), industrial and resource production (Appendix SS-C), transportation (Appendix SS-D), and waste disposal (Appendix SS-E). Additional statistics on socioeconomic information, government policies, and global warming potentials were also collected from published data of National Statistics Bureau (Appendix SS-F, SS-G, SS-H). All appendix can be accessed in Zenodo depository (Zhang, 2024). For transportation derived emission, the yearly volume of national oil consumption was applied due to lack of provincial data. All dynamic emission factors were retrieved from previous reported values, and assumed consistent across all provinces. All activity datasets were collected during the period of 2015-2019, which may reflect the effects of air pollution mitigation practices implemented since the introduction of “Clean Air Act”.

Line 109-110: The detailed algorithm of the bottom-up emission inventories was provided in full details in the Supporting Information, with activity level data and dynamic emission factors uploaded in data repository (Zhang, 2023).

The following are changes we made in the supporting information:

Line 26-28: Supplementary spreadsheet SS-A1 provided the vanadium content in raw coal produced in each coal producing province, which was calculated by associating the average vanadium concentration with the coal transmission matrix (Wu et al., 2020), and the results were reported in previous work (Liu et al., 2018).

Line 36-38: We have adopted the application rates of APCDs during 2015-2019 in equation S1-1, assuming the technology penetration of APCDs remained unchanged during 2015-2019, given the short study period and variability of number of APCDs yearly installation.

Line 39-40: SS-A5 provided the yearly consumed volume of raw coal at provincial level, with data sourced from the Yearbook of Coal Industry of China 2015-2019 (National Bureau of Statistics of China, 2020a).

Line 48-49: SS-A6 presented the provincial coke consumption level during 2015-2019, with data provided by China Coal Industry Statistical Yearbook 1949-2021' (National Bureau of Statistics of China, 2024a).

Line 68-69: Data results from China Energy Statistical Yearbook 1986-2023 was utilized (National Bureau of Statistics of China, 2024b), providing the yearly provincial consumption of oil products.

Line 89-92: SS-C2, C3, C4 and C5 provided the activity levels of each province in year 2015-2019 in sectors of steelmaking, glass manufacturing, coal mining and oil extraction, respectively, with datasets sourced from statistical yearbooks released by National Bureau of Statistics of China (National Bureau of Statistics of China, 2023a; National Bureau of Statistics of China, 2024a; National Bureau of Statistics of China, 2024b; National Bureau of Statistics of China, 2024c).

Line 124-126: The activity level of vanadium mining was provided in SS-C7, with data sourced from China Mining Sector Statistical Yearbook 2002-2019 (National Bureau of Statistics of China, 2024d).

Line 153-154: SS-E3 provided the yearly volume of incinerated wastes throughout 2015-2019, with data provided by China Urban Rural Construction Statistical Yearbook 2006-2022 (National Bureau of Statistics of China, 2023c)

Line 157-159: SS-E4 provided the amount of MSW admitted to landfill during 2015-2019, with data provided by China Urban Rural Construction Statistical Yearbook 2006-2022 (National Bureau of Statistics of China, 2023c).

Line 169-171: SS-E6 provided the provincial level wastewater discharge during 2015-2019, with data provided by China Urban Rural Construction Statistical Yearbook 2006-2022 (National Bureau of Statistics of China, 2023c).

We have provided the references for the data source in the Supporting Information:

National Bureau of Statistics of China: China Steel Industry Statistical Yearbook 1985-2022 [Data set], https://www.shujuku.org/china-steel-industry-yearbook.html, 2023a.

National Bureau of Statistics of China: China Automobile Industry Yearbook 1986-2022 [Data set], https://www.shujuku.org/china-automobile-industry-yearbook.html, 2023b.

National Bureau of Statistics of China: China Urban Rural Construction Statistical Yearbook 2006-2022 [Data set], https://www.shujuku.org/china-urban-rural-construction-statistical-yearbook.html, 2023c.

National Bureau of Statistics of China: China Coal Industry Statistical Yearbook 1949-2021 [Data set], https://www.shujuku.org/china-coal-industry-yearbook.html, 2024a.

National Bureau of Statistics of China: China Energy Statistical Yearbook 1986-2023 [Data set], https://www.shujuku.org/china-energy-statistical-yearbook.html, 2024b.

National Bureau of Statistics of China: China Light Industry Statistical Yearbook 1989-2022 [Data set], https://www.shujuku.org/china-light-industry-yearbook.html, 2024c.

National Bureau of Statistics of China: China Mining Sector Statistical Yearbook 2002-2019 [Data set], https://www.shujuku.org/mining-yearbook.html, 2024d.

Wu, B., Tian, H., Hao, Y., Liu, S., Sun, Y., Bai, X., Liu, W., Lin, S., Zhu, C., Hao, J., Luo, L., Zhao, S., and Guo, Z.: Refined assessment of size-fractioned particulate matter (PM2.5/PM10/PMtotal) emissions from coal-fired power plants in China, Science of The Total Environment, 706, 135735, https://doi.org/10.1016/j.scitotenv.2019.135735, 2020.

Yuan L. Vanadium in Coal Mining Area: Distribution, Modes of Occurrence, and Environmental Behavior. University of Science and Technology of China; Heifei, China: 2018.

2) In Figure 1, the arrows going into the atmosphere compartment sum up 205.466t. is there something missing in there, or is the reported 211.094t wrong? Could the authors clarify it, please? Also, if needed, make it clear in the text.

Reply: We greatly thank you for pointing out this issue. We have identified the cause of the mistake. For flux arrows of atmospheric emission, we showed the average emission values (1318 t and 78 t) for the industrial process by mistake, rather than the cumulative values from the five years study period (6595 t and 393 t). We have double checked all other values in the figure, and made sure all values shown here are consistently using cumulative values. After revision, the final atmospheric emission is 211 095 t.

Figure 1: Environmental flux model for vanadium directly discharged from anthropogenic activities into environmental receptors. Vanadium flux data was calculated cumulatively during 2015-2019 from different anthropogenic sectors with vanadium emission, i.e., coal combustion, stationary source oil burning, transportation source, industrial processes and waste disposal, all of which were labeled by numbers. Lines with arrow head indicated the vanadium flux from different sources to respective receptors, with line thickness proportional to the magnitude of vanadium flux.

In addition, we also updated the figures in the manuscript texts.

Line 11-13: Cumulative flux model reveals emissions of 211095 t, 3725 t, and 0.1 t of vanadium into atmosphere, soil and water, respectively. Coal combustion (46.7%) and stationary oil burning (40.1%) are the largest vanadium contributors.

3) Are there any studies with vanadium measurements to corroborate with the inventory emission findings? I mean, are there evidences that those environmental fluxes numbers make sense? Perhaps, comparison with previous studies or observed measurements?

Reply: Thank you for the question. We have compared our inventory with the global emission inventory for 2015–2016. In both cases, vanadium emissions were predominantly released into the atmosphere, resulting in a significant atmospheric emission flux. This pattern aligns closely with our findings. Additionally, the average yearly output during 2015–2019 accounted for 6% of the estimated annual global atmospheric output. This value suggests that our inventory reasonably captures China's contribution to atmospheric vanadium emissions, as it aligns with the scale of China's fossil fuel consumption and industrial activities in comparison to global data.

Line 129-132: Based on a global inventory developed during 2015-2016, 730000 t of vanadium was released into environment in a year due to anthropogenic activity (Schlesinger et al., 2017). As a major energy consumer and industrial developed nation, China’s yearly emission is estimated to comprise 6% of global emission output based on comparison with the global result.

Line 135-136: Globally, 82.2% of vanadium was directly released into atmosphere during 2015-2016 (Schlesinger et al., 2017), with 54.6% and 24.6% of emission from combustion of oil products and coal, respectively.

Line 138-140: In comparison, global inventory assumed all mining and industrial production as the major source of emission flux to the soil (Schlesinger et al., 2017), with a relative contribution of 18.5%.

Line 141-151: Both China and global inventory highlighted the dominance of atmospheric flux. However, while coal combustion was the predominant source for atmospheric emission in China, combustion of oil products weighed more importantly for global inventory. There was also a marked difference in estimated emission flux to soil, with a greater soil flux due to global industrial and mining activities. In contrast, the presented inventory showed a much smaller proportion of vanadium flux to soil receptor, because it was assumed that the majority of emission, from the large contributors such as steelmaking and glass production, was released into atmosphere. The development of global inventory employed mainly the available data of industrial activities in the United States, which lacked sufficient data resolution for developing nations. However, coal combustion played more important roles in China’s energy sector than petroleum. The contrast could also be attributed to variation in vanadium releasing fraction, emission reducing technology, vanadium content in raw materials in different regions, as well as assumptions made during inventory development. To improve the inventory, it is essential to perform local level investigation to increase the spatial and temporal specificity.

4) Could the authors expand on the use (or the lack) of technologies to reduce/control the vanadium emissions in the different sectors? Which technologies were in use during that period and what would be expected for more recent years emissions? Are there new technologies being applied in the last years, especially on the coal and oil burning sectors? Also, some info about the regional and national laws on technology control methods. For example, do they change at different provinces/regions? Could that help to explain the differences?

Reply: Thank you for the question. We have provided the application rates of end-of-pipe treatment technologies, mainly for the coal related emission. The source of database comprised primarily of statistics from the China Electricity Councils, the Permit Management System, and reported values in literature studies. Data were organized into a chart (Appendix SS-A3-2). Since 2010s (12^th Five Year Plan), national plan actions have aimed to reduce fine particulate matter (PM2.5) emission for industrial emissions, promoting the adoption of baghouse filter. In addition, the deployment of wet flue gas desulfurization has increased since 2005. In our study, we calculated the application rates of this two APCDs using yearly report data for newly commissioned facilities. For other conventional treatment system, extrapolation was carried out to calculate their application rates based on previous studies, as their usage has been in decreasing trend during 2000s. It was assumed that the application rates remain nearly constant during 2015-2019, given the relatively short study period, and the variability of numbers of newly commissioned system. Moreover, for NO_x removal, increasingly more facilities installed selective catalytic reduction system, which may involve the use of V₂O₅ as catalyzers, thereby potentially increasing the vanadium emission. We suspected that the massive utilizing of catalyzers may offset some of the reducing effects of APDCs on vanadium emission control. In regions with more advanced economy, such as EC, SC, and NC, the installation rates of APCDs were higher due to stricter enforcement of environmental regulations. We have incorporated this discussion into the manuscript to provide a more comprehensive understanding of APCD application rates and their implications for vanadium emission control.

Line 232-244: The large-scale installation of air pollution control devices (APCDs) commenced since 2004 (Wang et al., 2020e). After the introduction of “Clean Air Act”, the energy and industrial manufacturing sectors underwent a decreasing trend in pollutant emission (e.g., SO2, NOx) (Zheng et al., 2018), and transportation emission (e.g., NOx, CO, NMVOC) remained relatively steady. Despite these achievements, the overall activity levels continued to rise, along with vanadium output. The implementation of ultra-low emission (ULE) standards mandated the retrofitting of power plants with more efficient air pollutant control devices (APCDs), driving widespread technological upgrade. According to APCDs statistics (Appendix: SS-A-3), electrostatic precipitators have remained prevalent during study period due to the cost effectiveness. In comparison, a remarkable increase in application rates of baghouse filters and wet flue gas desulfurization was observed compared to previous report (Liu et al., 2015). Moreover, catalytic reduction systems for NOx removal were increasingly incorporated into newly commissioned facilities (Liu et al., 2019). However, catalytic reduction process often involved vanadium-based catalyst, potentially contributing to an increase in atmospheric emission. Despite advance in APCD’s implementation, the growing activity level driven by elevated energy consumption and demand for vanadium in various sectors may offset these improvements, underscoring the complex interplay between technological progress and industrial expansion.

Line 265-266: Additionally, according to statistics (Appendix SS-A3), NC has one of the highest application rates of APCDs including the advanced baghouse filters, which has the highest capture efficiency for fine particles.

We also added the following explanation to detail our source and handling for APCDs data in Supporting Information.

Line 30-40: SS-A3 provided the vanadium removal efficiencies of typical air pollution control devices (APCDs) and the spatial variation in application rates of these treatment systems. For electrostatic separator, baghouse filter, and wet method flue gas desulfurization, technology penetrate rates of these treatment systems in newly commissioned or upgraded coal combustion facility were calculated based on available data from China Electricity Council (China Electricity Council, 2024). For old-fashion wet scrubber and cyclone separators, it was assumed that their installation rates decreased. The application rates of these systems were extrapolated based on previously reported values (Liu et al., 2011). We have adopted the application rates of APCDs during 2015-2019 in equation S1-1, assuming the technology penetration of APCDs remained unchanged during 2015-2019.

Appendix SS-A3-2.

5) Concerning the spatial Vanadium releases in the 7 geographic regions, I think it would enrich the discussion if the authors include and relate socioeconomic characteristics and the findings. For example, it would be terrific to see if there is any correlation between GWP or other socioeconomic indexes (population, number of industries, income, …) and specific V emission sectors. I think the discussion in this 3.3 manuscript section does not represent the complexity observed in such a huge country with strong spatial discrepancies as China.

Reply: We appreciate your insightful comment. The socioeconomic and global warming potentials are indeed important determinants influencing the vanadium emission inventories. We have retrieved these datasets from the national statistic bureau for 2015-2019, including GDP, income, urban population, industry scale (Appendix SS-F), and other data in relation with global warming, such as thermal power generation (SS-H1), CO₂ emission (SS-H2), and utilization of cleaner energy (natural gas) (SS-H3). We have expanded the discussion on how socioeconomic factors affected the emission trends across different regions. The results showed that it is more likely for economic advanced region to implement energy reform, switching coal to natural gas, and increase the consumption of petroleum. For energy producing area with less developed economy, technological upgrade still plays critical role in further reducing its emission contribution. 

Line 166-171: Unlike the developed nations, China remained heavily reliant on coal, accounting for 51% of world annual coal consumption (Wang et al., 2020d). This dependency was further compounded by the continued expansion of installed power generation capacity (Wang et al., 2020e).

Line 258-266: In NC, Shanxi (420 t) and Inner Mongolia (358 t) experienced the largest increases in coal derived emission. According to Energy Development Strategy Actional Plan 2014-2020, both provinces were designated as major bases of large-scale thermal power generation, contributing significantly to the national energy supply network through long-distance electricity transport. Both provinces have faced economic challenges with one of the lowest GDP growth and income rises in China (Appendix SS-F), making energy substitution unlikely. In contrast, Beijing and its surrounding areas have achieved substantial reductions in coal use through Regional Collaboration on Air Pollution Control, aiming to reduce 25% of PM2.5 concentrations in Beijing-Tianjin-Hebei by phasing out small coal-fired boilers and switching to natural gas (Yan et al., 2018). Additionally, according to statistics (Appendix SS-A3), NC has one of the highest application rates of APCDs including the advanced baghouse filters, which has the highest capture efficiency for fine particles.

Line 276-279: SWC region significantly reduced coal consumption while maintained a steady growth of power generation (Appendix SS-H). This pattern coincided with increased investment on shale gas supply, particularly in Sichuan province with one of the most natural gas consumptions in China.

Line 279-281: Similarly, in delta area (Shanghai-Jiangsu-Zhejiang) of Yangtze River, the most economically developed region with the highest GDP and income per capita (Appendix SS-F), a significant decline in emission related to coal burning was observed, which aligned with a rapid increase in natural gas consumption (Appendix SS-H3).

Line 300-303: However, the emission in Hebei experienced a decreasing trend. Since the introduction of the Steel Industry Adjustment and Upgrading Plan (2016), the elimination of inefficient production processes decreased the generation of slag and dusts, the consolidation of industry into larger enterprise further encouraged the recycling and re-utilization of solid waste, such as converting slag into construction materials or recovering vanadium.

Line 308-310: The National Main Functional Zone Plan (2010) explicitly designated the western region of China as a base for energy resource development and heavy industry. Local governments in the western region have introduced investment incentives to attract energy-intensive enterprises to establish operations, and ease the environmental burden of eastern regions.

Line 313-320: The policy changes usually imposed more direct impact on coal consumption. In coastal region with strong economy and public awareness, the enforcement of environmental policies has become very urgent because wealthier populations tend to be more concerned with environmental well-being, and these regions have more financial and technological resources for implementing environmental policies. For example, coastal region made huge investment on liquified natural gas terminals and processing facilities, which can support the steady growth in natural gas usage. However, for poorer provinces, there is less incentive to prioritize environmental concerns, as the immediate focus is typically on supporting economic development. These regions may rely on cheaper energy sources, such as coal, to fuel industrial growth and meet the energy demands of expanding economies.

6) In the uncertainty section the authors applied a Pearson correlation as a sensitivity analysis. The reason for such and the conclusions from that are unclear. What did the authors intend to show through those correlations? What did they expect as results? Please expand the discussion on that.

Reply: Thank you for the question. Sensitivity analysis evaluates how variations in input parameters influence the outputs of a model. In this study, we tried to identify which emission source groups most significantly impact total emissions. To enhance the clarity, we have improved the Figure 4 analysis by providing more details on the linear relationship between individual input data profile and inventory output, measured by Pearson’s correlation coefficient. The stronger correlation (much closer to 1 or -1) indicate the potential source of uncertainties, specifically for inventories exhibiting high level of uncertainties. This insight underscores the need to focus on improving the reliability of these specific input datasets. To better interpret the findings, we expanded our discussion to include how sensitivity analysis delineates the linear relationships between individual input parameters and the inventory outputs. Strong correlations for parameters linked to highly uncertain outputs suggest priority areas for action, such as refining measurement methodologies or enhancing data resolution. These improvements will ultimately reduce the uncertainties in emission inventories and increase the robustness of the overall database.

Figure 4. Sensitivity analysis of input parameters using Pearson rank correlation coefficient. The strength and direction of linear relationship between subgroup emission and input parameters related to (A) consumption level, (B) production level, and (C) other emission factors, were measured using Pearson correlation method (abbreviation was used as following: ESP – electrostatic precipitator, FB – baghouse filter, CS – cyclone separator, WS – wet scrubber, WFGD = wet method flue gas desulfurization, PCFB – pulverized coal fired boiler, CGB – chain-grate boiler, FBCB – fluidized bed combustion, CO – coke oven).

Line 121-124: For sensitivity study, the linear relationship between input parameters and emission levels were evaluated using Pearson rank correlation coefficient, ranging from -1 to 1. Any input parameter with the absolute coefficient values closer to 1 indicate strong correlation, which may suggest the source of uncertainties (Liu et al., 2021).

Line 332-347: Pearson’s correlation coefficients were further computed to measure the linear relationship between input parameters and emission inventories. For subgroup inventories with high level of uncertainties, transportation related oil consumption and wastewater sludge generation were significantly correlated (p < 0.05) with emission levels (Figure 4A&4B). Both sectors lacked activity level data at provincial level, resulting in imprecise or insufficient data resolution for inventory development. For coal combustion, vanadium release fraction of typical boilers exhibited significant correlation (p < 0.05) with emission level. Some release fraction data were retrieved from old studies from other countries, which may not well-represent the overall status in China. Vanadium content in raw coal and coke was another influential factor (p < 0.05), as its high variability may introduce significant uncertainty in the emissions calculation. The application rates of electrostatic precipitator, bag filter, and wet method of flue gas desulfurization showed stronger correlation with emission inventory, which may serve as potential source of uncertainties. It was also challenging to access the temporal and spatial data on technology penetration for APCDs. The data profile assumed that application rates remained unchanged during the study period, and extrapolation was made for old-fashioned technologies (e.g., electrostatic precipitator) based on previous report (Liu et al., 2011). To reduce the uncertainty level, more focus should be allocated to improve the data resolution of transportation related activity levels. It is also vital to perform on-site measurement for domestic facility process in raw coal and coke combustion sectors in order to improve the data quality of emission factors. For example, previous study employed continuous emission monitoring system to provide real-time tracking on emission for computation of emission factor (Tang et al., 2020).

7) Finally, a significative discussion is missing around the Vanadium emissions compared to other locations and studies. I would like to encourage the authors to put the Chinese Vanadium emissions into a wider perspective to the reader. That is, include some discussion comparing the emissions in China with other locations/countries. For example, in the research of Bai et al. (2021) V emissions were estimated till the year 2017 for China, however, the authors did not compare the results with those findings. What about other counties?

Reply: Thank you. We also agreed it is important to compare the inventory with other study the difference between inventory development procedure and the source of contrast. We have added another section 3.5 to specifically discuss the difference between this study and the previous work done by Bai et al. In addition, we have added more discussion our inventory against the global context. In addition, we have also expanded comparison with global trend of vanadium emission.

Line 164-171: Globally, the annual vanadium emission from coal combustion in 2016 was estimated to be only one fourth of the output in the previous decades (Schlinger et al., 2017). This reduction was largely due to the emission reduction measures implemented in developed nations, such as the United States and Western Europe since 1980s (Environmental Health & Engineering, Inc., 2011; Arienzo et al., 2021). Unlike the developed nations, China remained heavily reliant on coal, accounting for 51% of world annual coal consumption (Wang et al., 2020d). This dependency was further compounded by the continued expansion of installed power generation capacity (Wang et al., 2020e). Additionally, the average vanadium concentrations in China’s raw coal were higher than the global average levels (Bartoňová et al., 2023), resulting in a disproportionately large contribution to global vanadium output and led to more challenges associated with domestic emission.

Line 178-180: Compared with inventory derived from global oil consumption, the annual vanadium output in China accounted for smaller proportion (~6.1%) of global vanadium emission in 2016 (Schlesinger et al., 2017), which has also increased by 2.5 folds since 2000 (Monakhov et al., 2004).

Line 183-186: The reduction in fuel oil derived emission also aligned with the overall decline in global demand (International Energy Agency, 2016), likely attributed to the substitution of residual oil with use of natural gas in refinery activities (Visschedijk et al., 2013). This trend suggested that regulations in various countries prioritized limiting emissions from vanadium rich heavy oil.

Line 197-203: For heavy duty vehicles, sulfur emission remained nearly unchanged despite increased logistic activities (Zheng et al., 2018). This trend can be attributed to the strict regulation of vehicle diesel standard (GB 19146) introduced after 2013, which in turn reduced the demand in vanadium-based catalyzer (Bai et al., 2021). In maritime transportation, fuel oil was primarily utilized by shipping vessels. Although global vanadium emissions from maritime transportation decreased due to sulfur restrictions (IMO MARPOL Annex VI) imposed by International Maritime Organization in the early 21st century (Arienzo et al., 2021), China's inventory showed a progressive increase in vanadium output. This suggested that the expanded shipping activities has offset the effects of stricter sulfur emission controls.

Line 207-220: In the recent decade, China undertook strong action to reduce the overcapacity in steel industry, resulting in a slow production growth (Zhu et al., 2022). The implementation of ultralow emission transformation (ULET) markedly reduced emission of PM2.5 and other pollutants (Tang et al., 2020). However, vanadium emission continued to rise, making up of 85% consumption in industrial production sector (ResearchInChina, 2018). The new standard for high strength rebar required more vanadium for alloying process (Polyak, 2019). In contrast, emission from mining and oil exploration in China did not show significant growth, consisting with the global trend (Schlesinger et al., 2017). The relative contribution to China emission inventory from coal mining (2.5%) and oil extraction (2.2%) were significantly lower than global inventory, where oil exploration and mining activity accounted for 56% and 18% of emission output, respectively (Schlesinger et al., 2017). Such contrast could be explained by higher level of data resolution in China’s inventory, capturing a more detailed scope of mining and oil extraction activities, whereas the global inventory used more generalized data and assumptions to fill the data gap, leading to a broader, less precise estimates. For vanadium ore mining, the overall vanadium production remained stable in major producing powers such as Russia and South Africa (Polyak, 2015; Polyak, 2017; Polyak, 2019). However, the newly adopted environmental policy placed a ban on importation of vanadium slags in China (Polyak, 2019), which may stimulate the demand for domestic vanadium production.

Line 227-228: Recycling rate of steel slags in China was at only 20-30%, compared to a much higher utilization rate of 97% in developed nations (Yang et al., 2024).

Line 355-369: In a previous study on historical trend of vanadium emission in China (Bai et al., 2021), the emission from coal combustion significantly declined following a peak in 2007, with emission levels in 2010s lower than this study. This contrast may largely stem from differences in collection and handling of APCDs data. According to China Electricity Council, major technological upgrades since 2010s led to a sharp increase in implementation of baghouse filter and wet method desulfurization. However, we assumed that their annual installation rates have stabled in recent years. Notably, the study on vanadium removal efficiencies of baghouse filter (79.0 ± 17.9%) was extremely scarce and may be seriously undermined compared to conventional electrostatic precipitators (89.7 ± 13.7%). In the previous study, the relative contribution of both stationary and mobile source showed a decreasing trend until 2017. In comparison, oil consumption data in this inventory showed a persistently increasing trend in all sectors except for fuel oil and diesel. For transportation emission, our study utilized the amount of fuel consumption reported by national statistics bureau, whereas vehicle counts were integrated into the emission calculation in previous study. It should be noted that the substantial rise in electrical and natural gas-powered vehicle after 2015 may affect the emission calculation. Therefore, difference in data source may contribute to the discrepancy between inventories. Moreover, both studies agreed on significant increase in emission pertinent to industrial production and solid waste disposal, highlighting the contribution of steel and glass production to the rise of vanadium output. However, the present study also underscored the importance of solid waste disposal, which were neglected in the contemporary study.

Specific points:

1) P1, L19: “... Emissions pertinent to raw coal ad coke combustion was...” were, not was. Also, please add the word “respectively” at the endo f this sentence, if appropriate.

Reply: Thank you. We have revised the sentences accordingly.

Line 20-24: Inventory uncertainties were driven by insufficiently resolved activity data, poorly characterized emission factors, and variability in vanadium content in raw materials, particularly in coal combustion, transportation, and waste disposal (e.g. sludge disposal).

2) P1, L27: ... accounts for...

Reply: Thank you.

Line 30-31: As the largest vanadium producer, Sichuan and Hebei in China account for vast majority of vanadium output, with principal vanadium forms in vanadium-titanium deposits (Yu et al., 2015)

3) P3, L70: Could you please cite the figures in here?

Reply: Thank you. We have cited the table here.

Line 72-73: The environmental impact was assessed utilizing the life cycle assessment (LCA) that accounted for five major anthropogenic sources leading to vanadium emission (Table 1).

4) P4, L86: Could you detail the unit of this total production volume (EPi)? Is it in tons, or m3, ...?

Reply: Thank you. We have specified the unit of the total production volume.

Line 96-99: Where 𝐸𝑉 was the vanadium emission in metric ton, 𝐸𝐹𝑗 was the emission factor in kg vanadium emitted per ton of products; A𝑃𝑗 was the total production amounts of industrial products (or wastes) in metric ton.

5) P6, L136: The raw coal and coke presented in Figure 1 sum up 100.372t, however, the Figure S2 displays a total amount of 100.381t. is there a reason for such inconsistency?

Reply: Thank you for pointing out these errors. We have forgotten updating the display of calculation result in Figure S2 after revisions.

The following are our response to comment 5-8: The total emission volume associated with coal and coke combustion is 100372 t. For oil burning, transportation, and waste, the correct figures are 86038 t, 17649 t, and 3277 t, respectively. We also corrected the labels for Transportation and Industrial processes, which were mis-placed. The percentage of each category was also added accordingly.

6) ? Which one is correct? Note that also oil burning (86.001t vs 86.068t), transportation (17.649t vs 17650t) and waste (3.277t vs 3279t) emissions do not match. Please verify.

Reply: Thank you. Please see our response in comment 5.

7) Figure S2: I believe the industrial process and Transportation labels are missplaced. Please verify.

Reply: Thank you. Please see our response in comment 5.

8) Figure S2: Could the authors please add the percentages (%) of each categorie in this figure? Those % are mentioned in the text ~P5, however, they were not displayed. Please make sure they are in line with the #5-6 above.

Reply: Thank you. Please see our response in comment 5.

9) P10, L225-226: This new sentence is not making sense, please rewrite or clarify.

Reply: Thank you. We have improved the expression of this sentence.

Line 289-291: The steel industry was the primary source of emission inventory related to industrial production, with the largest emission from Hebei province (1587.2 t) due to concentrated steel making operations.

10) P10, L227: Please correct the typo analyzed.

Reply: Thank you.

Line 325: The uncertainties of input data were firstly evaluated by bootstrapping simulation.

In addition, we have incorporated more references to support our discussion:

Line 398: Arienzo, M. M., Legrand, M., Preunkert, S., Stohl, A., Chellman, N., Eckhardt, S., Gleason, K. E., and McConnell, J. R.: Alpine Ice‐Core Evidence of a Large Increase in Vanadium and Molybdenum Pollution in Western Europe During the 20th Century, JGR Atmospheres, 126, e2020JD033211, https://doi.org/10.1029/2020JD033211, 2021.

Line 404: Bartoňová, L., Raclavská, H., and Najser, J.: Vanadium – Valuable and toxic element in coal combustion ash: An overview, Process Safety and Environmental Protection, 172, 923–940, https://doi.org/10.1016/j.psep.2023.02.070, 2023.

Line 408: Emission of Hazardous Air Pollutants from Coal-Fired Power Plants, Environmental Health & Engineering, Inc., www.obamawhitehouse.archives.gov/sites/default/files/omb/assets/oira_2060, 2011.

Line 424: International Energy Agency: World Energy Statistics 2016, OECD, https://doi.org/10.1787/9789264263079-en, 2016.

Line 427: Liu, K., Wu, Q., Wang, L., Wang, S., Liu, T., Ding, D., Tang, Y., Li, G., Tian, H., Duan, L., Wang, X., Fu, X., Feng, X., and Hao, J.: Measure-Specific Effectiveness of Air Pollution Control on China’s Atmospheric Mercury Concentration and Deposition during 2013–2017, Environ. Sci. Technol., 53, 8938–8946, https://doi.org/10.1021/acs.est.9b02428, 2019.

Line 430: Liu, C., Zhang, L., Wen, Y., and Shi, K.: Sensitivity analysis of O3 formation to its precursors-Multifractal approach, Atmospheric Environment, 251, 118275, https://doi.org/10.1016/j.atmosenv.2021.118275, 2021.

Line 432: Monakhov, I. N., Khromov, S. V., Chernousov, P. I., and Yusfin, Yu. S.: The Flow of Vanadium-Bearing Materials in Industry, Metallurgist, 48, 381–385, https://doi.org/10.1023/B:MELL.0000048420.68839.2a, 2004.

Line 436: ResearchInChina: Global and China Vanadium Industry Report 2018-2023, http://www.researchinchina.com/Htmls/Report/2018/10513.html, 2018.

Line 440: Polyak, D.E.: 2015 Minerals Yearbook, Vanadium, U.S. Geological Survey, https://d9-wret.s3.us-west-2.amazonaws.com/assets/palladium/production/mineral-pubs/vanadium/myb1-2015-vanad.pdf, 2016.

Line 442: Polyak, D.E.: 2017 Minerals Yearbook, Vanadium, U.S. Geological Survey, https://d9-wret.s3-us-west-2.amazonaws.com/assets/palladium/production/atoms/files/myb1-2017-vanad.pdf, 2020.

Line 445: Polyak, D.E.: 2019 Minerals Yearbook, Vanadium, U.S. Geological Survey, https://pubs.usgs.gov/myb/vol1/2019/myb1-2019-vanadium.pdf, 2024.

Line 456: Tang, L., Xue, X., Jia, M., Jing, H., Wang, T., Zhen, R., Huang, M., Tian, J., Guo, J., Li, L., Bo, X., and Wang, S.: Iron and steel industry emissions and contribution to the air quality in China, Atmospheric Environment, 237, 117668, https://doi.org/10.1016/j.atmosenv.2020.117668, 2020.

Line 467: Visschedijk, A. H. J., Denier Van Der Gon, H. A. C., Hulskotte, J. H. J., and Quass, U.: Anthropogenic Vanadium emissions to air and ambient air concentrations in North-West Europe, E3S Web of Conferences, 1, 03004, https://doi.org/10.1051/e3sconf/20130103004, 2013.

Line 481: Wang, Q., Song, X., and Liu, Y.: China’s coal consumption in a globalizing world: Insights from Multi-Regional Input-Output and structural decomposition analysis, Science of The Total Environment, 711, 134790, https://doi.org/10.1016/j.scitotenv.2019.134790, 2020d.

Line 484: Wang, G., Deng, J., Zhang, Y., Zhang, Q., Duan, L., Hao, J., and Jiang, J.: Air pollutant emissions from coal-fired power plants in China over the past two decades, Science of The Total Environment, 741, 140326, https://doi.org/10.1016/j.scitotenv.2020.140326, 2020e.

Line 500: Yan, D., Lei, Y., Shi, Y., Zhu, Q., Li, L., and Zhang, Z.: Evolution of the spatiotemporal pattern of PM2.5 concentrations in China – A case study from the Beijing-Tianjin-Hebei region, Atmospheric Environment, 183, 225–233, https://doi.org/10.1016/j.atmosenv.2018.03.041, 2018.

Line 503: Yang, M. and Yang, J.: Vanadium extraction from steel slag: Generation, recycling and management, Environmental Pollution, 343, 123126, https://doi.org/10.1016/j.envpol.2023.123126, 2024.

Line 531: Zhu, S., Gao, C., Song, K., Gao, W., Guo, Y., and Gao, C.: The changes in spatial layout of steel industry in China and associated pollutant emissions: A case of SO2, Journal of Environmental Management, 302, 114034, https://doi.org/10.1016/j.jenvman.2021.114034, 2022.