Historical and projected trends of long-term nitrogen dynamics from China’s mariculture Authors: Jingyi Liu, Tomás Marín Del Valle, Long Xu, Benhui Zhu, Hanle Xu, Jianguo Liu, Wu Yang Published: 2026 ABSTRACT Mariculture sustains global food security yet generates substantial nitrogen waste through uneaten feed and metabolic excretion, with poorly quantified long-term fluxes. We assessed nitrogen dynamics of China’s mariculture from 1983 to 2024 and projected future trends under 12 scenarios. Here we show that nitrogen discharge surged 150-fold to 93.73–128.37 kilotons in 2024, driven by fed species. Concurrently, nitrogen removal by shellfish and seaweed rose from 5.06–8.80 kilotons to 89.03–137.68 kilotons. Spatially, southern provinces emerge as dominant discharge hotspots, whereas northern provinces demonstrate strengthened sink potential. Scenario analyses reveal a critical tipping point: although currently a net sink at −9.36 kilotons, the sector risks reverting to a net source by 2030 under most trajectories, potentially peaking at 15.14 kilotons. Strategic interventions, including expanding cultivation areas and increasing density for both fed and extractive species, could expand net removal to 100.47 and 245.71 kilotons by 2060 and 2100, respectively. Introduction To ensure the sustainability of global food systems and mitigate environmental impacts for a growing population, it is crucial to understand the nutrient discharge impacts of food production systems, particularly for nitrogen1,2. However, terrestrial agriculture struggles to meet escalating protein demand amid finite land resources and competing land-use pressures3–5. In response, mariculture is emerging as an alternative source of protein6,7. FAO (Food and Agriculture Organization of the United Nations) data underscore this shift. The global production of mariculture animal products has surged from 9.2 million tons annually in the 1990s to 35.3 million tons in 2022, highlighting its increasingly relevant role in food systems8. In comparison to land-based protein products, mariculture products are generally perceived as having lower negative environmental externalities and superior nutritional profiles9–11. Furthermore, when effectively managed, mariculture has the potential to mitigate fishing pressure on wild marine stocks3,12,13. These advantages not only offer innovative solutions for food security but also substantially contribute to achieving multiple Sustainable Development Goals (SDGs), including Zero Hunger (SDG 2), Climate Action (SDG 13), and Life Below Water (SDG 14)14,15. While the rapid expansion of mariculture presents promising solutions for food security, it also brings forth significant environmental challenges16,17. These include destruction of natural habitats18, the introduction of microplastics19 and antibiotics20, and improper discharge of aquaculture wastewater21, all of which contribute to significant pollution and the misallocation of valuable resources22,23. Among these concerns, nitrogen discharge stands out as a particularly crucial issue, particularly originating from intensive fed mariculture (e.g., finfish and crustaceans in open cages or ponds)24. Unlike extractive species, these fed systems rely on external protein inputs, where substantial nitrogen enters the marine environment through two primary pathways: particulate organic nitrogen from uneaten feed and feces, and dissolved inorganic nitrogen from metabolic excretion25,26. Although nitrogen plays an essential role as a nutrient within aquaculture ecosystems, excessive accumulation can lead to water quality deterioration27. The unchecked accumulation of nitrogen can trigger toxic algal blooms, which not only degrade water quality but also threaten aquatic life and public health, resulting in dire ecological and economic consequences28,29. As the world’s largest mariculture producer, China has seen its industry grow substantially along its 18,000 km coastline, accounting for ~60% of global mariculture output in 20228. In response to the rising demand for seafood, the nation has proactively implemented a suite of policies aimed at diversifying cultured species and promoting advanced cultivation techniques, such as deep-sea cage mariculture30. Uniquely, this rapidly developing sector is characterized by a dominance of non-fed (extractive) species, with algae and bivalves constituting most production volume, thereby playing a critical role in marine nutrient cycling and carbon sequestration. The rapid expansion not only contributed to economic development and food security but also intensified concerns regarding nitrogen dynamics and their environmental impacts31. The nitrogen dynamics within mariculture systems exhibit species-specific characteristics. In open-exchange production systems, such as nearshore cages and flow-through ponds, the predominant nitrogen efflux pathways originate from unconsumed feed particles and metabolic wastes32. These nutrient loads are subsequently mobilized through water exchange, ultimately discharging dissolved and particulate nitrogen fractions into adjacent marine ecosystems33–35. Conversely, non-feeding species such as filter-feeding bivalves and seaweed can serve as biological agents for nutrient removal, thus mitigating nutrient loads within the ecosystem31,36,37. In alignment with China’s commitment to carbon neutrality by 2060 and its robust environmental protection policies, understanding nitrogen dynamics from its mariculture sector is paramount for achieving sustainable development and safeguarding ecosystem health. Despite widespread recognition of nitrogen pollution as a critical issue in coastal zones, long-term comprehensive quantification and spatial analysis of nitrogen discharge from China’s extensive mariculture industry is largely missing38. Existing research revealed the dual impacts of mariculture. Fed culture types (including cage culture and pond culture) are significant contributors to nutrient release, while extractive culture types (e.g. seaweed, filter-feeding shellfish culture) act as substantial nutrient sinks31. For instance, shrimp and fish ponds in Hainan discharged an estimated 612 tons of total dissolved nitrogen annually into coastal waters, contributing to coastal eutrophication and impacting seagrasses and corals39. Extractive mariculture, particularly seaweed cultivation, demonstrates a considerable nitrogen uptake capacity, with estimates ranging from ~70.62 to 75.00 kilotons of nitrogen absorbed annually from the Chinese coasts40,41. Zhang et al. quantified the nitrogen budget of China’s mariculture in 2019, estimating a nitrogen discharge of 58.45 kilotons predominantly from fed culture types and a removal of 109.25 kilotons via extractive culture types31. Some other studies quantified province-level emissions, reporting that mariculture fish generated nitrogen emissions of 35.92 kilotons in Fujian Province and 41.77 kilotons in Guangdong Province in 2020. In contrast, seaweed farming in Fujian alone absorbed around 58.44 kilotons of nitrogen annually37. Given the rapid growth of mariculture, a critical deficiency in existing research is the focus on short-term ( < 5 years) nitrogen balance, with a notable scarcity of long-term accounting at national and provincial levels, particularly across diverse mariculture cultivated species31,42. Furthermore, the absence of projections of future nitrogen dynamics under varied socioeconomic and technological scenarios largely hinders effective management of discharge regulation. This study aims to fill this gap by conducting a comprehensive assessment of nitrogen discharge and removal from China’s mariculture from 1983 to 2024. Focusing on the ten coastal provinces of Mainland China, we estimated nitrogen discharges and removals for different cultivated species at both national and subnational levels (Fig. 1). Furthermore, we developed predictive models using the STIRPAT (Stochastic Impacts by Regression on Population, Affluence, and Technology) framework to project future nitrogen discharge and removal under 12 development scenarios up to 2100 43,44. By analyzing historical trends and potential future pathways, our research provides actionable insights for policymakers and industry stakeholders to optimize nitrogen management strategies for China’s mariculture sector and beyond.Fig. 1Spatiotemporal trends of mariculture area (103 ha) and yield (104 t) in China’s coastal provinces (1983–2024).a Geographic location of the study area. b, c Changes in mariculture area for fed (b) and extractive (c) species. d, e Changes in yield for fed (d) species and extractive (e) species. Spatiotemporal trends of mariculture area (103 ha) and yield (104 t) in China’s coastal provinces (1983–2024). a Geographic location of the study area. b, c Changes in mariculture area for fed (b) and extractive (c) species. d, e Changes in yield for fed (d) species and extractive (e) species. Results Long-term national trends of nitrogen dynamics The primary contributors to nitrogen discharge within China’s mariculture were the cultivation of fish, shrimp, crab, and sea cucumber (Fig. 2). Between 1983 and 2024, the national total nitrogen discharge from these species increased by ~150-fold, rising from 0.59–1.07 kilotons to 93.73–128.37 kilotons (Fig. 3a). Throughout the study period, shrimp mariculture emerged as the predominant contributor to nitrogen discharge, followed by fish, crab, and sea cucumber. Notably, over the past two decades, the contribution of fish has exhibited a consistent upward trajectory. By 2024, fish mariculture accounted for 51.77% of the total discharge, followed by shrimp (37.10%), crab (11.04%), and sea cucumber (0.09%), respectively. In the meantime, the nitrogen removal by shellfish and seaweed cultivation also experienced rapid growth, escalating from 5.06–8.80 kilotons in 1983 to 89.03–137.68 kilotons in 2024 (Fig. 3a). Seaweed cultivation was the dominant contributor (nearly 70%) to total nitrogen removal, with an annual growth rate of 6.80%.Fig. 2Temporal trends and spatial patterns of nitrogen discharge and removal in China’s six major mariculture categories (1983–2024).a Total nitrogen discharge and removal of six categories over the 42-year period (104 t), with the overall average annual growth rate. b–g Category-specific nitrogen discharge trends (104 t), highlighting regional disparities across the ten coastal provinces. Vector icons (e.g., fish, tiger prawn, crab, sea cucumber) in panel a are courtesy of the Integration and Application Network (IAN) Symbol Library (ian.umces.edu), licensed under CC BY-SA 4.0 (https://creativecommons.org/licenses/by-sa/4.0/deed.en). Minor formatting adjustments were made to align with the journal style.Fig. 3Temporal patterns of nitrogen discharge and removal from six major mariculture products (fish, shrimp, crab, sea cucumber, shellfish, and seaweed) from 1983 to 2024.a National total. b–k Ten coastal provinces. Temporal trends and spatial patterns of nitrogen discharge and removal in China’s six major mariculture categories (1983–2024). a Total nitrogen discharge and removal of six categories over the 42-year period (104 t), with the overall average annual growth rate. b–g Category-specific nitrogen discharge trends (104 t), highlighting regional disparities across the ten coastal provinces. Vector icons (e.g., fish, tiger prawn, crab, sea cucumber) in panel a are courtesy of the Integration and Application Network (IAN) Symbol Library (ian.umces.edu), licensed under CC BY-SA 4.0 (https://creativecommons.org/licenses/by-sa/4.0/deed.en). Minor formatting adjustments were made to align with the journal style. Temporal patterns of nitrogen discharge and removal from six major mariculture products (fish, shrimp, crab, sea cucumber, shellfish, and seaweed) from 1983 to 2024. a National total. b–k Ten coastal provinces. Synthesizing these discharge and removal dynamics, the trend of net nitrogen discharge from China’s mariculture between 1983 and 2024 demonstrates a complex pattern, initially declining and then fluctuating in the last 20 years (Fig. 3a). A critical observation is that since the beginning of the 21st century, China’s mariculture sector has likely functioned as a net nitrogen sink (Fig. 3a). Exceptions were observed in 2003 and 2018—notably associated with a surge in shrimp discharge—where median net discharge values turned positive. This signals a potential shift toward net nitrogen source status during these years (Fig. 3a). Furthermore, over the last four decades, the annual growth rate of nitrogen discharge has been 12.89%, significantly outpacing the annual nitrogen removal rate of 7.14%. Spatiotemporal patterns of historical nitrogen Our analyses reveal a spatially complementary effect: southern provinces emerged as dominant nitrogen discharge sources, whereas northern provinces functioned as sustained ecological sinks between 1983 and 2024 (Figs. 3 and 4). As visually highlighted in Fig. 4, a distinct latitudinal divergence has formed over the decades, with the coastline clearly divided into blue sinks in the north and orange sources in the south (Supplementary Figs. 1–2). Specifically, Guangdong (482.82–709.63 kilotons) and Fujian (367.14–443.31 kilotons) in southern China exhibited the largest cumulative nitrogen discharges—collectively accounting for ~50% of the national total (Fig. 3h, i), with Guangxi showing the highest growth rate (17.30%) (Fig. 3j). Conversely, northern provinces such as Liaoning and Shandong Provinces demonstrated a consistent and accelerating capacity for nitrogen removal via large-scale extractive farming, with cumulative removal amounts of 414.68–618.87 kilotons and 673.42–1004.86 kilotons, respectively (Fig. 3d, e). Guangdong contributed the largest cumulative net discharge (387.39–657.80 kilotons), followed by Guangxi (98.46–142.60 kilotons) and Jiangsu (35.55–177.70 kilotons). Both Tianjin and Zhejiang Provinces peaked in net nitrogen discharge ~2000, but their subsequent trajectories diverged dramatically (Fig. 3b, g). Situated in the transitional zone, Zhejiang Province has successfully transitioned from a net emitter to a net remover, exemplifying the potential for regional restructuring. Meanwhile, Hebei and Jiangsu Provinces displayed considerably greater variability in their nitrogen dynamics, underscoring the instability of their current systems (Fig. 3c, f).Fig. 4Spatial distribution patterns of net nitrogen discharge from China’s mariculture across coastal provinces from 1983 to 2024.The maps illustrate the spatiotemporal evolution of net nitrogen fluxes (104t) for the selected years 1983, 1990, 2000, 2010, 2015, and 2024. Spatial distribution patterns of net nitrogen discharge from China’s mariculture across coastal provinces from 1983 to 2024. The maps illustrate the spatiotemporal evolution of net nitrogen fluxes (104t) for the selected years 1983, 1990, 2000, 2010, 2015, and 2024. The spatial patterns of net nitrogen dynamics were closely linked to mariculture structural composition (Fig. 3b–k). While shrimp cultivation has emerged as the predominant source of nitrogen discharge across most provinces in China (contributing ~50% of total discharge), regional disparities remain significant. Southern provinces like Fujian, Guangdong, and Hainan represent exceptions where fish farming takes precedence. In contrast, sea cucumbers, which were introduced to the mariculture sector relatively recently, have begun to play a significant role in nitrogen discharge, particularly in the northern provinces of Hebei, Liaoning, and Shandong, as well as in the southern province of Fujian. Notably, in Fujian Province, since the inception of sea cucumber cultivation in 2003, the associated nitrogen discharge has markedly increased, exhibiting an annual average growth rate of 71.77%. These findings highlight the dynamic and region-specific characteristics of nitrogen discharge associated with various mariculture species in China. Scenario-based national projections of nitrogen The STIRPAT model results indicate that nitrogen discharge from China’s mariculture is primarily driven by technological advancement (ln T, represented by the number of aquatic technology extension organizations), mariculture scale (ln S), and density (ln D). Notably, scale (ln S) acts as the dominant factor, with a 1% increase associated with a 0.75% rise in discharge (Table 1). In terms of future projections, discharge under scenarios S1.1 and S2.1 is projected to decline moderately by 0.16% and 3.04% by 2100, respectively. Conversely, under the other baseline scenarios (S3.1, S4.1, and S5.1), discharge is expected to rise by 1.68% to 11.08% compared to the 2024 level. Under restrictive development scenarios (S1.2, S2.3, and S3.2), substantial reductions are anticipated, with discharge dropping to nearly half the level in the 2024 baseline by 2100. By contrast, under mariculture expansion scenarios—particularly S5.2, characterized by scale expansion coupled with density increase—nitrogen discharge is projected to surge, reaching 3.42 times that of 2024 by 2100.Table 1Results of regression models for predicting future nitrogen discharge and removal from China’s maricultureVariableln NDln NRCoefficientSECoefficientSEΔln P–3.438*1.6760.9581.776ln A0.0630.096–0.0740.134ln T0.211***0.062–0.0620.069ln D0.583***0.0520.896***0.067ln S0.752***0.0741.185***0.112Constant–7.287***0.585–7.413***0.725Obs325259R2within=0.840; overall=0.785within=0.841; overall=0.850TestF (5,310) = 325.37***F (5,246) = 260.47***Notes: ln ND is derived from panel data of ten coastal provinces from 1983 to 2015, while ln NR is based on data from eight coastal provinces (excluding Tianjin and Guangxi) for the same period. *P < 0.05, ***P < 0.001. Table 1Results of regression models for predicting future nitrogen discharge and removal from China’s maricultureVariableln NDln NRCoefficientSECoefficientSEΔln P–3.438*1.6760.9581.776ln A0.0630.096–0.0740.134ln T0.211***0.062–0.0620.069ln D0.583***0.0520.896***0.067ln S0.752***0.0741.185***0.112Constant–7.287***0.585–7.413***0.725Obs325259R2within=0.840; overall=0.785within=0.841; overall=0.850TestF (5,310) = 325.37***F (5,246) = 260.47***Notes: ln ND is derived from panel data of ten coastal provinces from 1983 to 2015, while ln NR is based on data from eight coastal provinces (excluding Tianjin and Guangxi) for the same period. *P < 0.05, ***P < 0.001. Results of regression models for predicting future nitrogen discharge and removal from China’s mariculture Notes: ln ND is derived from panel data of ten coastal provinces from 1983 to 2015, while ln NR is based on data from eight coastal provinces (excluding Tianjin and Guangxi) for the same period. *P < 0.05, ***P < 0.001. Furthermore, regression analysis reveals that nitrogen removal from mariculture is significantly driven by mariculture density (ln D) and mariculture scale (ln S), with mariculture scale exhibiting the strongest positive effect (Table 1). In contrast, population growth (Δln P) and affluence (ln A) show no statistically significant association with removal (Table 1). Under scenarios S4.2 and S5.2—characterized by high-scale expansion and continued intensive practices for both fed and extractive species—nitrogen removal continues to rise throughout the 21st century. By 2100, removal volumes reach 590.96 and 571.68 kilotons, respectively, corresponding to a 4.18-fold and 4.01-fold increase relative to 2024 level. In contrast, under the majority of the other scenarios where cultivation scale is constrained, or density is reduced, the nitrogen removal capacity is projected to contract significantly, with reductions ranging from 21.30% to 68.06% by 2100 compared to the 2024 levels. Net nitrogen discharge from China’s mariculture—encompassing both fed species (fish, crustaceans) and extractive species (shellfish, seaweed)—is projected to increase sharply under most scenarios. However, this trajectory could be reversed under pathways driven by sustainable intensification and structural optimization (Fig. 5). All baseline scenarios (S1.1, S2.1, S3.1, S4.1, and S5.1) exhibit a critical transition from a net nitrogen sink (−9.36 kilotons in 2024) to a net nitrogen source by 2100. This functional reversal is most pronounced in scenarios S5.1, which projects a net discharge of 33.33 kilotons, followed by S1.1 (28.53 kilotons), S4.1 (24.79 kilotons), S2.1 (23.46 kilotons), and S3.1 (16.67 kilotons). In contrast, S1.3, S4.2, and S5.2 scenarios are characterized by a trajectory of rapid accumulation. Specifically, these pathways reflect a synchronized intensification pattern, where the stocking densities of both nitrogen-emitting fed species and nitrogen-removing extractive species increase concurrently. Under these pathways, the sector not only maintains its role as a nitrogen sink but substantially amplifies its removal capacity. Starting from the baseline of 9.36 kilotons in 2024, net nitrogen removal is projected to experience a multifold surge, reaching 135.82 kilotons (S1.3), 245.71 kilotons (S4.2), and 213.22 kilotons (S5.2) by 2100, respectively.Fig. 5Predictions of net nitrogen discharge of mariculture under 12 scenarios.a National total. b–k Ten coastal provinces. Predictions of net nitrogen discharge of mariculture under 12 scenarios. a National total. b–k Ten coastal provinces. Provincial disparities in projected nitrogen trends Provincial trajectories of net nitrogen discharge diverge sharply under different development pathways (Fig. 5). Specifically, Guangxi Province consistently emerges as the largest net source, whereas Liaoning Province acts as a significant net sink across all scenarios. In terms of nitrogen discharge, Shandong Province ranks first, accounting for over 20% of the national total in all scenarios; however, a distinct downward trend in its discharge is observed under S1.2, S2.3, and S3.2 scenarios (Fig. 5; Supplementary Fig. 3). Conversely, regarding nitrogen removal, Shandong and Liaoning Provinces emerge as the predominant regions. Although Fujian Province surpasses Liaoning Province to rank second in S1.2, S2.3, and S3.2 scenarios, the combined contribution of Shandong and Liaoning Provinces remains robust, consistently exceeding 50% of the national total. This dominance culminates in S4.2 and S5.2 scenarios, where their aggregate nitrogen removal reaches a peak of 66.11% (Supplementary Fig. 4). The projected changes in net nitrogen discharge across the ten provinces reveal a clear divergence in trajectories under different development pathways (Fig. 6). Among the five SSP baseline scenarios, S3.1 demonstrates the most moderate projected increase in net nitrogen discharge from 2030 to 2100, with variations of 68.33% ± 86.07% (mean ± SD) in the short term, 72.85% ± 93.89% in the mid-term, and 54.99% ± 69.03% in the long term. Conversely, among all twelve scenarios, the S4.2 scenario projects the most substantial increase in the short term and mid-term, with net discharge rising by 104.92% ± 127.46% and 237.54% ± 283.64%, respectively. While the S5.2 scenario has the largest net nitrogen discharge long-term changes, with increases of 342.48% ± 466.83%.Fig. 6Projected changes in nitrogen discharge and removal of China’s mariculture under 12 scenarios from 2030 to 2100 relative to the reference year of 2024.The boxplots show the median (center line), upper and lower quartiles (box limits), and 5th and 95th percentiles (whiskers). The short-term, mid-term and long-term refer to the years 2030, 2060, and 2100, respectively. Projected changes in nitrogen discharge and removal of China’s mariculture under 12 scenarios from 2030 to 2100 relative to the reference year of 2024. The boxplots show the median (center line), upper and lower quartiles (box limits), and 5th and 95th percentiles (whiskers). The short-term, mid-term and long-term refer to the years 2030, 2060, and 2100, respectively. Discussion This study presents a comprehensive, multi-decadal analysis of nitrogen dynamics in China’s mariculture sector, integrating historical trends (1983–2024) with future projections under a range of plausible development scenarios. By quantifying the contributions of key mariculture species to nitrogen discharge and removal, we establish a critical foundation for rethinking the sustainability of mariculture. Our findings reveal a stark imbalance: nitrogen discharge has surged by 150-fold, while removal has merely expanded by 16-fold. This disproportionate growth underscores the extensive expansion of the industry and the escalating pressure exerted on the environment. Crucially, we demonstrate that this trajectory is not inevitable. Through scenario-based modeling using the Shared Socioeconomic Pathways (SSPs), we demonstrate that—contrary to the assumption that environmental recovery requires reducing production—transitioning from a net nitrogen emitter to a net remover by mid-century (~2040) entails expanding cultivation scale and increasing stocking densities for both fed and extractive species. This co-intensification paradigm leverages a macro-level ecological offset: by strategically coupling the two, the substantial biomass and nutrient-assimilation capacity of extractive species scales up to fully offset, and ultimately surpass, the nitrogen discharge from the intensified fed sector. The spatial distribution of nitrogen discharge and removal across China’s coastal provinces exhibits marked heterogeneity, shaped by climatic conditions, species-specific requirements, and regional cultivation practices31,45. Southern provinces such as Fujian and Guangdong currently dominate discharge due to the prevalence of fed aquaculture in subtropical waters. In contrast, northern provinces—including Shandong and Liaoning—provide ideal conditions for nitrogen-removing species such as shellfish and seaweed, benefiting from cold, nutrient-rich waters and lower temperatures37,46. Future development should not be driven solely by production potential but by ecological function. Specifically, mitigating discharge risks in southern aggregation zones necessitates a multi-pronged approach: the widespread adoption of Integrated Multi-Trophic Aquaculture to assimilate dissolved nitrogen47; and the deployment of precision feeding technologies to minimize waste at the source48. While such spatial heterogeneity highlights the limitations of one-size-fits-all policies, the implementation of these region-specific strategies must be orchestrated within a unified national framework to prevent fragmented governance. We propose a macro-level coordination mechanism, such as an inter-provincial ecological compensation system or nitrogen trading scheme, to ensure the co-intensification of fed and extractive mariculture49,50. Specifically, provinces with high demand for fed species but limited environmental carrying capacity could offset their nitrogen discharge by providing financial or technological support to sink-dominant provinces (e.g., Shandong and Fujian) to further expand their extractive mariculture51,52. This market-driven and state-coordinated mechanism would ensure that while regional management remains highly tailored to local conditions, the aggregate national trajectory remains steered toward a net nitrogen sink, effectively balancing food security with environmental sustainability. Future projections under the SSPs framework—incorporating key socioeconomic drivers, specifically population dynamics and GDP growth—reveal a critical divergence in trajectories. Under business-as-usual scenarios—particularly SSP5 (S5.1), the sector risks reverting to a significant net nitrogen source (33.33 kilotons by 2100). However, our multi-scenario analysis reveals a critical dynamic: expanding cultivation scale and increasing stocking density for both fed and extractive species are not inherently detrimental but are prerequisite drivers for achieving a net sink status (e.g., reaching net removal of 245.71 kilotons in 2100 under S4.2 scenario). This indicates that the scale effect of mariculture can be harnessed for environmental benefits. By leveraging the substantial biomass base of China’s shellfish and seaweed sectors, the absolute nitrogen removal capacity scales up sufficiently to outpace discharge inputs from intensified fed aquaculture, and even represents a potential offset of ~5% of the total national nitrogen input53. Consequently, these predictive scenarios empirically validate the aforementioned co-intensification paradigm. It confirms that sustainability is achieved through strategic expansion rather than degrowth, effectively scaling the localized benefits of ecological engineering into a national-level nitrogen mitigation mechanism supported by governance frameworks like Marine Functional Zoning54. In conclusion, this study advances the scientific understanding of nitrogen dynamics in mariculture by providing a comprehensive, long-term, spatially explicit, and scenario-based assessment of its environmental trajectory. It not only quantifies the scale of the challenge but also demonstrates that sustainable mariculture is not only feasible but achievable through coordinated, ecologically informed development. Our analyses focus on organism-level nitrogen uptake and release and do not explicitly resolve the subsequent biogeochemical fate of nitrogen in coastal waters, such as sedimentation or denitrification, which warrants further integration with coupled hydrodynamic–biogeochemical models. By aligning mariculture expansion with ocean conservation, China’s mariculture industry has the potential to establish a global paradigm for sustainable blue food systems, where economic development and ecological preservation are harmonized—transforming the traditional trade-off between production intensity and marine ecosystem vitality into a synergistic relationship. Methods Data sources This study focuses on the ten coastal provinces of mainland China—Liaoning, Hebei, Tianjin, Shanghai, Jiangsu, Zhejiang, Fujian, Guangdong, Guangxi, and Hainan-which collectively account for over 95% of the nation’s mariculture production. We compiled a comprehensive panel dataset from 1983 to 2024 using the China Fisheries Statistical Yearbook (1984–2025)55, the China Statistical Yearbook56, and provincial-level yearbooks. Key variables included mariculture product output and area, provincial population, GDP per capita, and the number of aquatic technology extension organizations. The nitrogen discharge parameters for fish, crustaceans (shrimp and crabs), and shellfish, as well as the nitrogen content parameters for shellfish and seaweed, were sourced from previous studies (Supplementary Tables 1–4). Future projections of provincial population and GDP data under the SSPs were derived from a dataset released by the Science Data Bank57, ensuring consistency with global scenario frameworks (Supplementary Figs. 5–7). Estimation of mariculture nitrogen dynamics Nitrogen discharge and removal from China’s mariculture were estimated using a species-specific, process-based framework that distinguishes between nitrogen-emitting (feed-based) and nitrogen-removing (extractive) species58. For feed-based species—including fish, crustaceans (shrimp and crabs), and sea cucumbers—nitrogen discharge was calculated using the Emission Factor method, developed by IPCC (Intergovernmental Panel on Climate Change)31. These factors encompass various forms of nitrogen, including ammonia nitrogen (NH₃-N), nitrate nitrogen (NO₃−-N), nitrite nitrogen (NO₂−-N), and organic nitrogen59. For nitrogen-removing species—primarily filter-feeding shellfish and seaweed—the removal was estimated using biomass-based absorption coefficients. Following the approach of Xiong et al. (2023), the nitrogen removal by seaweed was calculated based on the net nutrients assimilated into harvested biomass and long-term sequestered organic matter. Nitrogen released back into the water column during growth (e.g., via tissue erosion) is excluded from this calculation to accurately represent the net sink capacity37. For shellfish species, we used the following formula to estimate the amount of nitrogen they remove by transferring suspended particles in water to benthic sediments31:1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${F}_{{{\rm{R}}},\,i}={r}_{i}\times \left[{t}_{i}\times {C}_{{{\rm{m}}},i}+(1-{t}_{i})\times {C}_{{{\rm{s}}},i}\right]/100$$\end{document}FR,i=ri×ti×Cm,i+(1−ti)×Cs,i/100where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${F}_{{{\rm{R}}}}$$\end{document}FR (t/t) represents the nitrogen removal factor; r is the shellfish’s dry and wet weight ratio; t is the ratio of dry tissue weight to total dry weight; \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${C}_{{{\rm{m}}}}$$\end{document}Cm and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${C}_{{{\rm{s}}}}\,$$\end{document}Cs(%) are N contents in dry tissue and shell respectively; i denotes a specific species within the shellfish category (See Supplementary Tables 2 and 4 for details).2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${N}_{\,{{\rm{sh}}}}=\mathop{\sum }\limits_{i}({F}_{i}-{F}_{{{\rm{R}}},i})\times {M}_{i}$$\end{document}Nsh=∑i(Fi−FR,i)×Miwhere \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${N}_{{{\rm{sh}}}}$$\end{document}Nsh (104 t) represents the nitrogen discharge from shellfish mariculture products in a province; \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${F}$$\end{document}F (t/t) represents nitrogen discharge factors of main mariculture products (Supplementary Table 1); M (104 t) represents the yield of species. We used the following formula to calculate nitrogen discharge and removal by province: For \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${N}_{{{\rm{sh}}}}$$\end{document}Nsh ≥ 03\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${N}_{{{\rm{D}}}}={N}_{{{\rm{a}}}}+{N}_{{{\rm{sh}}}}=\mathop{\sum }\limits_{j}{F}_{j}\times {M}_{j}+{N}_{{{\rm{sh}}}}$$\end{document}ND=Na+Nsh=∑jFj×Mj+Nsh4\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${N}_{{{\rm{R}}}}={N}_{{{\rm{SW}}}}={\mathop{\sum }\limits_{k}}{F}_{{{\rm{R}}},k}\times {M}_{k}$$\end{document}NR=NSW=∑kFR,k×Mk For \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${N}_{{{\rm{sh}}}}$$\end{document}Nsh <05\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${N}_{{{\rm{D}}}}={N}_{{{\rm{a}}}}={\mathop{\sum }\limits_{j}}{F}_{j}\times {M}_{j}$$\end{document}ND=Na=∑jFj×Mj6\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${N}_{{{\rm{R}}}}=\left(-{N}_{s{{\rm{h}}}}\right)+{N}_{{{\rm{SW}}}}=\left(-{N}_{{{\rm{sh}}}}\right)+{\mathop{\sum }\limits_{k}}{F}_{{{\rm{R}}},k}\times {M}_{k}$$\end{document}NR=−Nsh+NSW=−Nsh+∑kFR,k×Mkwhere \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${N}_{{{\rm{D}}}}$$\end{document}ND and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${N}_{{{\rm{R}}}}$$\end{document}NR (104 t) represent the total nitrogen discharge and removal for a province, respectively; \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${N}_{{{\rm{a}}}}$$\end{document}Na (104 t) is the amount of nitrogen discharged from mariculture of fish, shrimp, crab, and sea cucumber in a province; \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${N}_{{{\rm{SW}}}}$$\end{document}NSW (104 t) represents the amount of nitrogen removed by seaweed mariculture in a province; j represents the respective species among fish, shrimp, crabs, and sea cucumbers. Additionally, k corresponds to the various species within the seaweed category. When \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${N}_{{{\rm{sh}}}}\ge 0$$\end{document}Nsh≥0, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${N}_{{{\rm{D}}}}$$\end{document}ND is calculated as the sum of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${N}_{{{\rm{a}}}}$$\end{document}Na and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${N}_{{{\rm{sh}}}}$$\end{document}Nsh, while \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${N}_{{{\rm{R}}}}$$\end{document}NR is equivalent to \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${N}_{{{\rm{SW}}}}$$\end{document}NSW. Conversely, if \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${N}_{{{\rm{sh}}}} < 0$$\end{document}Nsh<0, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${N}_{{{\rm{D}}}}$$\end{document}ND is equivalent to \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${N}_{{{\rm{a}}}}$$\end{document}Na, with \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${N}_{{{\rm{R}}}}$$\end{document}NR represented by the combined \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$-{N}_{{{\rm{sh}}}}$$\end{document}−Nsh and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${N}_{{{\rm{SW}}}}$$\end{document}NSW. The net nitrogen discharge for each province was then computed as the difference between total discharge and total removal, with the sign indicating whether the province acts as a net emitter (positive) or net remover (negative).7\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${N}_{{\mbox{n}}}={N}_{{\mbox{D}}}-{N}_{{\mbox{R}}}$$\end{document}Nn=ND−NRwhere \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${N}_{{\mbox{n}}}$$\end{document}Nn (104 t) represents net nitrogen discharge from mariculture in a province. Given that a small portion of the production of fish, shrimp, crabs, shellfish, and seaweed in mariculture remains unspecified in the statistical yearbook, we adopted a quantile-based uncertainty analysis method to estimate discharge or removal factors for this subset of products. Specifically, we designated the upper quartile of the discharge or removal factors within each category as the upper limit for the discharge or removal factors of these products, the median as the central value, and the lower quartile as the lower limit. This methodology enabled us to establish a range for nitrogen discharge and removal in China’s mariculture sector. The maximum net discharge was derived under the condition of maximum discharge and minimum nitrogen removal; the minimum net discharge was derived under the condition of minimum discharge and maximum nitrogen removal. Methods for predicting the future trend Future nitrogen dynamics in China’s mariculture were projected using an extended STIRPAT model, which integrates socioeconomic drivers with ecological processes to simulate long-term trajectories under diverse development pathways60. The STIRPAT framework allows for the incorporation of additional driving factors based on specific research needs and empirical relevance. In this study, the dependent variables (I) were the total nitrogen discharge (ND) and total nitrogen removal (NR) from mariculture in each province. We incorporated five independent variables as drivers for ND and NR (Table 2), reflecting key influences on mariculture activities—population (P), GDP per capita (A), the number of aquaculture technology extension stations (T), mariculture density (D), and mariculture area (S). Specifically, the variable T serves as a proxy for the technological capacity available for diffusion and adoption at the grassroots level. As the primary agents responsible for transferring environmentally friendly practices to farmers in China, these extension organizations play a decisive role in enhancing the environmental efficiency of production. The models were specified as follows:8\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\mathrm{ln}}{\,N}_{{\mbox{D}}}={{\mathrm{ln}}\,a+\alpha }_{1}\varDelta {\mathrm{ln}}\,P+{\alpha }_{2}{\mathrm{ln}}\,A+{\alpha }_{3}{\mathrm{ln}}\,T+{\alpha }_{4}{\mathrm{ln}}\,D+{\alpha }_{5}{\mathrm{ln}}\,S+{\mathrm{ln}}\,\varepsilon$$\end{document}lnND=lna+α1ΔlnP+α2lnA+α3lnT+α4lnD+α5lnS+lnε9\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\mathrm{ln}}{\,N}_{{\mbox{R}}}={\mathrm{ln}}\,b+{\beta }_{1}\varDelta {\mathrm{ln}}\,P+{\beta }_{2}{\mathrm{ln}}\,A+{\beta }_{3}{\mathrm{ln}}\,T+{\beta }_{4}{\mathrm{ln}}\,D+{\beta }_{5}{\mathrm{ln}}\,S+{\mathrm{ln}}\,\delta$$\end{document}lnNR=lnb+β1ΔlnP+β2lnA+β3lnT+β4lnD+β5lnS+lnδwhere \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\mathrm{ln}}\,a$$\end{document}lna and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\mathrm{ln}}\,b$$\end{document}lnb are the model parameters, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\alpha }_{1}$$\end{document}α1 to \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\alpha }_{5}$$\end{document}α5 and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\beta }_{1}$$\end{document}β1 to \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\beta }_{5}$$\end{document}β5 are the elasticities terms for independent variables, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\mathrm{ln}}\,\varepsilon$$\end{document}lnε and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\mathrm{ln}}\,\delta$$\end{document}lnδ are the error terms of the models.Table 2Description of variables used in the construction of STIRPAT models for the projection of nitrogen discharge and removal from mariculture in each coastal provinceVariableSymbolUnitDescriptionNitrogen dischargeND104 tTotal nitrogen discharge of the provinceNitrogen removalNR104 tTotal nitrogen removal of the provincePopulationΔP%Annual percentage change in provincial populationAffluenceAyuanGross domestic product per capita of the provinceTechnological levelTnumberNumber of aquatic technology extension organizations in the provinceMariculture densityDtons per hectareMariculture yield per unit in the provinceMariculture scaleS103 hectaresMariculture area in the provinceThese variables were derived from 10 coastal provinces from 1983 to 2024. For variable T, data from 1983–2002 refers to aquatic technology extension stations, and from 2003–2024, it refers to aquatic technology extension organizations. Table 2Description of variables used in the construction of STIRPAT models for the projection of nitrogen discharge and removal from mariculture in each coastal provinceVariableSymbolUnitDescriptionNitrogen dischargeND104 tTotal nitrogen discharge of the provinceNitrogen removalNR104 tTotal nitrogen removal of the provincePopulationΔP%Annual percentage change in provincial populationAffluenceAyuanGross domestic product per capita of the provinceTechnological levelTnumberNumber of aquatic technology extension organizations in the provinceMariculture densityDtons per hectareMariculture yield per unit in the provinceMariculture scaleS103 hectaresMariculture area in the provinceThese variables were derived from 10 coastal provinces from 1983 to 2024. For variable T, data from 1983–2002 refers to aquatic technology extension stations, and from 2003–2024, it refers to aquatic technology extension organizations. Description of variables used in the construction of STIRPAT models for the projection of nitrogen discharge and removal from mariculture in each coastal province These variables were derived from 10 coastal provinces from 1983 to 2024. For variable T, data from 1983–2002 refers to aquatic technology extension stations, and from 2003–2024, it refers to aquatic technology extension organizations. The model parameters were estimated using panel data from 10 coastal provinces spanning 1983 to 2015 (training dataset), with fixed effects selected via Hausman test to account for unobserved heterogeneity. To ensure the reliability of the model for long-term projections, we conducted a validity check using an independent dataset covering the period from 2020 to 2024 (validation dataset). The coefficients derived from the training dataset were applied to predict nitrogen discharge and removal for this validation period. The comparison between predicted and observed values demonstrates a robust fit (see Supplementary Tables 5–7, and Supplementary Figs. 11–13 for detailed validation results), confirming the model’s reliability for scenario-based forecasting. After establishing the model’s validity, we developed twelve future pathways by translating the five distinct SSPs into specific mariculture production modes to reflect diverse policy and developmental realities (Table 3). The parameterization of the STIRPAT model variables and five SSP scenarios are characterized as follows: SSP1 (Sustainability) is characterized as Ecological and Precision Mariculture, the industry shifts towards high-quality development where environmental limits are strictly enforced. The farming scale (S) is constrained or reduced and stocking density (D) is lowered to minimize ecological footprints, while technological investment (T, e.g., extension stations) expands rapidly to promote precision feeding. SSP2 (Middle of the Road) represents a Business-as-Usual mode, assuming mariculture continues its historical trajectory where future trends for T, D, and S follow their respective historical annual growth rates, reflecting a balance between moderate production growth and gradual technological improvement. In contrast, SSP3 (Regional Rivalry), defined as Extensive and Resource-Dependent Farming, is driven by food security concerns and trade barriers that prioritize maximizing domestic supply quantity; this leads to an aggressive expansion of farming area (S) and density (D) under relaxed environmental regulations, whereas technological diffusion (T) stagnates due to a lack of investment. Alternatively, SSP4 (Inequality) depicts Intensified Industrial Monoculture, where the sector becomes dominated by capital-rich enterprises utilizing high-density industrial systems; consequently, stocking density (D) increases significantly through technological intervention and technology (T) grows in elite sectors, but overall spatial expansion (S) is limited by land/sea use consolidation. Finally, SSP5 (Fossil-fueled Development) follows a High-Input High-Output Expansion trajectory fueled by rapid economic growth and global market demand, resulting in the simultaneous rapid growth of scale (S), density (D), and technology (T) via an energy-intensive model that relies on technological fixes rather than prevention to address environmental issues61.Table 3Scenario design and variable descriptionScenarioDescriptionTDSSSP1S1.1 BaselineStay at the 2024 levelStay at the 2024 levelStay at the 2024 levelS1.2 Limiting developmentChange at current paceStay at the 2024 levelDecrease at current paceS1.3 Nitrogen neutralizationChange at current paceIncreased at current paceStay at the 2023 levelSSP2S2.1 BaselineStay at the 2024 levelStay at the 2024 levelStay at the 2024 levelS2.2 Business as usualChange at current paceIncrease at current paceDecrease at current paceS2.3 Limiting developmentChange at current paceStay at the 2024 levelDecrease at current paceSSP3S3.1 BaselineStay at the 2024 levelStay at the 2024 levelStay at the 2024 levelS3.2 Limiting developmentChange at current paceStay at the 2024 levelDecrease at current paceSSP4S4.1 BaselineStay at the 2024 levelStay at the 2024 levelStay at the 2024 levelS4.2 Mariculture expansionIncrease and stabilize after 2030Increase at current paceIncrease at current paceSSP5S5.1 BaselineStay at the 2024 levelStay at the 2024 levelStay at the 2024 levelS5.2 Mariculture expansionIncrease and stabilize after 2030Increase at current paceIncrease at current paceChanges in A, P refer to SSP database57. The expansion of T and S were both restricted to a twofold increase relative to 2024. Additionally, a lower threshold of 10% of the 2024 level was set for T, while D was subject to a physical upper limit of 40 t/ha (Supplementary Figs. 5–10). For the specific baseline values and historical changing rates for each province, please refer to Supplementary Table 8. Table 3Scenario design and variable descriptionScenarioDescriptionTDSSSP1S1.1 BaselineStay at the 2024 levelStay at the 2024 levelStay at the 2024 levelS1.2 Limiting developmentChange at current paceStay at the 2024 levelDecrease at current paceS1.3 Nitrogen neutralizationChange at current paceIncreased at current paceStay at the 2023 levelSSP2S2.1 BaselineStay at the 2024 levelStay at the 2024 levelStay at the 2024 levelS2.2 Business as usualChange at current paceIncrease at current paceDecrease at current paceS2.3 Limiting developmentChange at current paceStay at the 2024 levelDecrease at current paceSSP3S3.1 BaselineStay at the 2024 levelStay at the 2024 levelStay at the 2024 levelS3.2 Limiting developmentChange at current paceStay at the 2024 levelDecrease at current paceSSP4S4.1 BaselineStay at the 2024 levelStay at the 2024 levelStay at the 2024 levelS4.2 Mariculture expansionIncrease and stabilize after 2030Increase at current paceIncrease at current paceSSP5S5.1 BaselineStay at the 2024 levelStay at the 2024 levelStay at the 2024 levelS5.2 Mariculture expansionIncrease and stabilize after 2030Increase at current paceIncrease at current paceChanges in A, P refer to SSP database57. The expansion of T and S were both restricted to a twofold increase relative to 2024. Additionally, a lower threshold of 10% of the 2024 level was set for T, while D was subject to a physical upper limit of 40 t/ha (Supplementary Figs. 5–10). For the specific baseline values and historical changing rates for each province, please refer to Supplementary Table 8. Scenario design and variable description Changes in A, P refer to SSP database57. The expansion of T and S were both restricted to a twofold increase relative to 2024. Additionally, a lower threshold of 10% of the 2024 level was set for T, while D was subject to a physical upper limit of 40 t/ha (Supplementary Figs. 5–10). For the specific baseline values and historical changing rates for each province, please refer to Supplementary Table 8. Provincial population and GDP projections under these SSPs were obtained from previous studies57. We used these projected provincial population (P) and GDP (A) data from the SSPs outcome datasets (Supplementary Figs. 5–7). For the remaining variables—specifically, the number of aquaculture technology extension organizations (T), the mariculture density (D), and the mariculture scale (S)—we designed future trends based on scenario assumptions (Table 3; Supplementary Figs. 8–10). Supplementary information Supplementary Information Peer Review file Supplementary Information Peer Review file