Probabilistic source apportionment and quantification of nitrate contamination in a karst aquifer system: Revealed by stable isotopic and hydrochemical proxies

Probabilistic source apportionment and quantification of nitrate contamination in a karst aquifer system: Revealed by stable isotopic and hydrochemical proxies | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Probabilistic source apportionment and quantification of nitrate contamination in a karst aquifer system: Revealed by stable isotopic and hydrochemical proxies Zupeng Wan, Baojun Liu, Wenwen Chen, Yingjie Chen, Xiaoyu Yan, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8680219/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 10 You are reading this latest preprint version Abstract Karst basins are characterized by high permeability and complex surface-subsurface hydrological connectivity which intensifies nitrogen migration. As a typical karst basin located in southern Guilin, the Lipu River basin faces environmental pressures from intensive agriculture and industrial activities; however, the contributions of different nitrate sources and the dominant nitrogen cycling processes have not been systematically characterized. In this study, NO 3 − stable isotopes (δ 15 N-NO 3 − and δ 18 O-NO 3 − , hydrochemical parameters, and a Bayesian isotope mixing model were integrated to identify nitrate sources and elucidate nitrogen cycling in a karst river basin in Lipu County, southwestern China. The results show that total nitrogen concentrations in river water were significantly higher in winter (7.09 ± 6.58 mg/L) than in summer (2.95 ± 1.48 mg/L). Concentrations of NO 3 − in river water during summer were significantly lower than those in winter and were also lower than groundwater concentrations in both seasons, indicating a strong dilution effect during the rainy season. Among different land-use types, groundwater NO 3 − concentrations in industrial areas were significantly higher than those in agricultural and residential areas. These spatiotemporal patterns suggest that nitrogen pollution control should prioritize elevated winter nitrogen loads in river water, as well as nitrate contamination in groundwater associated with tributaries and industrial zones. Isotopic signatures combined with modeling indicate that manure and sewage contributed 56.7 ± 8.2% of total nitrogen inputs. This integrated isotopic and hydrochemical approach elucidates nitrogen distribution and source contributions in karst basins, providing scientific support for targeted mitigation strategies to reduce nitrogen-related environmental and health risks. Bayesian isotope mixing model Nitrate source apportionment Karst aquifer Uncertainty quantification Environmental risk assessment Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 1. Introduction Pollution control strategies established for porous-media dominated basins are often inadequate for highly heterogeneous karst environments (Guo et al. 2019 ). When nitrogen pollution from industrial and agricultural development overlays the unique surface-subsurface hydrological structure of karst basins, pollutant transport deviates from traditional model predictions (P. Zhang et al. 2023 ). A highly permeable karst conduit network enables refractory organic matter in industrial wastewater and agricultural nitrogen to bypass the slow degradation process in the soil zone for rapid and deep transboundary transport (Güven et al. 2017 ). Instead of attenuating pollution, this geological feature may intensify its transformation and induce biotoxicity through complex physicochemical processes, with accelerated and amplified ecological effects (Wells et al. 2018). Existing studies have focused on nitrogen source apportioning in non-karst areas (Pandit et al. 2022). However, the rapid surface-subsurface exchange in karst systems limits the applicability of traditional watershed models (e.g, SWAT, HEC-HMS), which struggle to accurately characterize heterogeneous flow paths and nitrogen transport processes, thereby undermining the reliability of source apportionment analyses (Iriawan H et al.2021). Consequently, accurate quantification requires a dynamic traceability network integrating isotope techniques and hydrogeological technologies of nitrogen transport pathways in karst systems, identification of key pollution sources, and analysis of their spatiotemporal differentiation patterns. To address these challenges, this study targets the Lipu River basin, where high nitrate concentrations have been documented yet source contributions remain unquantified, employing dual nitrate isotopes coupled with hydrochemical parameters to trace nitrogen sources and transformation pathways in this representative karst system. Isotope and receptor models quantitatively determined industrial and domestic pollution source contributions, advancing from static source tracing to dynamic process analysis. Aquatic nitrogen exists primarily as NO 3 − , NO 2 − and NH 4 + . High mobility and proportions of NO 3 − in karst areas make it a key indicator of pollution transport (Zuo et al. 2023 ). In addition, high NO 3 − levels are associated with methemoglobinemia in humans and aquatic ecosystem degradation (Yan et al. 2023 ). NO 3 − pollution sources in karst basins include point and river sources. Point sources include mainly industrial parks and landfills. In addition, NO 3 − concentrations and sources are affected by various nitrogen transformations. For example, higher NH 4 + concentrations can promote nitrification (Ma et al. 2021 ) and inhibit NO 3 − removal by assimilation, leading to NO 3 − accumulation in karst basins. Excessive nitrogen fertilizer application in agricultural areas is typically considered the main source of NO 3 − in the basin (Wang et al. 2024 ), whereas point-source pollution is considered the main source of NO 3 − pollution in urbanized areas (Lai et al. 2023 ). However, studies have demonstrated that manure, sewage, and soil nitrogen are the major contributors to groundwater NO 3 − pollution in agricultural areas (Cui et al. 2023 ), suggesting that rapid urbanization and agricultural activities are altering nitrogen pollution sources and contributions in river basins. Integrating multi-source data with models facilitates precise tracing and quantitative assessment of the dynamic load of each pollution source under the dual drive of nature–human enabling the transition from fuzzy attribution to precise localization. NO 3 − sources largely reflect basin nitrogen sources (Iqbal et al. 2022 ; Schaefer et al. 2016 ). Linking land use types to cations and anions has been a traditional approach to identifying NO 3 − sources and nitrogen cycling processes (Biddau et al. 2023 ), The NO 3 − stable isotope method (δ 15 N-NO 3 − and δ 18 O-NO 3 − ) provides meaningful insights into NO 3 − sources and nitrogen conversion processes because different NO 3 − sources exhibit unique isotopic characteristics (Xue et al. 2019 ). The NO 3 − stable isotope method requires minimal auxiliary information, offers high accuracy and operational simplicity, and directly identifies of NO 3 − sources (Yang et al. 2024 ). Generally, δ 15 N-NO 3 − is the highest in M&S (manure and sewage) and δ 18 O-NO 3 − is the highest in AD (Atmospheric deposition), whereas 15 N-NO 3 − is the lowest in mineral fertilizers, followed by in SN (soil nitrogen), which facilitates tracing of NO 3 − sources in aquatic environments (Wu P et al. 2022 ). However, overlappingNO 3 − isotope signatures from different sources limit the accuracy of the dual-isotope approach for source apportionment, such as nitrogen fertilizer to soil nitrogen, soil nitrogen to manure and sewage, mixing of multiple NO 3 − sources, and NO 3 − isotope fractionation during biogeochemical cycles (Lin et al. 2019 ). Consequently, numerous studies have sought to identify NO 3 − sources and transformation processes by integrating NO 3 − isotopes with complementary tracers (e.g, δ 34 S-SO 4 2− , δ 11 B), hydrochemical indicators (e.g, Cl − , Na + ), environmental parameters (e.g, dissolved oxygen), and multivariate statistical methods (e.g, redundancy analysis) (Grechanik et al. 2024 ; Li et al. 2023 ). This study integrated a Bayesian isotope mixing model with NO 3 ⁻ stable isotope analysis. Additional covariates (hydrochemical parameters and land use types) enhanced source apportionment accuracy compared with traditional mixing models. The hydrological status of the southwestern karst basin is complex, with intense agricultural and human activities. Groundwater, rainfall, and river water are key sources of river recharge. Towns and farmlands are scattered around karst basin (Chang et al. 2024 ). Therefore, karst basins are typical areas of mixed land use (Yi et al. 2015 ). Frequent seasonal groundwater-river water exchange results in severe groundwater nitrogen pollution, with average nitrogen and NO 3 − concentrations reaching 24.35 and 15.15 mg/L (Boumaiza et al. 2022 ), respectively, posing a major threat to lake water quality and safety. However, few studies have focused on nitrogen sources, transformation, and contributions of nitrogen sources in karst basins, potentially masking dominant nitrogen sources and pollution characteristics in rivers and groundwater. To address these research gaps, we investigated inorganic nitrogen pollution in river water and groundwater in the Lipu River Basin within Lipu County, considering domestic sewage, industrial sewage, and other pollution sources. By integrating NO 3 − stable isotope (δ 15 N and δ 18 O) analysis with water chemistry data, meteorological parameters, environmental factors, and the Bayesian isotope mixing (SIAR) model, the objectives of this study are to: (1) clarify the distribution characteristics of DIN species under the influence of various discharge types in a typical karst area; (2) investigate the spatiotemporal distribution of nitrogen-containing pollutants in river water and groundwater and their correlations with comprehensive hydrochemical indicators; and (3) identify the sources of NO 3 − based on nitrogen distribution characteristics and the SIAR model. These findings provide a scientific basis for evaluating nitrogen dynamics and developing management strategies for nitrogen pollution control under various discharge scenarios in karst regions. 1 Materials and methods 2.1 Description of the study area The study area is located in the northeastern part of Guangxi Zhuang Autonomous Region, in southern part of Guilin City, with Lipu City as the main administrative division. It includes the western part of Pingle County, extending to the west banks of the Lijiang and Guijiang Rivers (110°12′6″–110°40′57″ east longitude, 24°20′54″–24°39′27″ north latitude). The climate of the study area is characterized by a long, hot, and rainy summer, a warm and occasionally cold winter, distinct summer and winter, an average monthly temperature of 18.9°C over many years, and an average annual rainfall of 1441.5 mm, mainly concentrated in April–August. The submerged aquifer in the area is 0.5–3.5 meters deep. The lithology of the area is composed mainly of clastic, carbonate, igneous, and loose rocks. The Quaternary Holocene Q 4 is located on the upper terrace, with pore aquifers and partial pressure. The northern part of the basin overlies the Lower Carboniferous stages of the Datangdian and Yanguandian C 1 , which contain fissure cave water with a small flow of springs (Zhang et al. 2020 ). 2.1.1 Watershed monitoring Concentrated sampling was conducted during the summer and winter seasons in 2022, as shown in Fig. 1 , Fourteen sampling points were selected along the mainstream of the Lipu River (uniformly distributed from the upstream Tianjingling to the downstream Guijiang confluence point), including 13 sampling points from the six major tributaries of the Pulu, Dumo, Datang, Xinping, Limu, and Maling Rivers. Twelve groundwater sampling points were selected based on the direction of groundwater movement and underlying hydrogeological conditions. 2.1.2 Sample collection Based on the hydrogeological conditions of the study area and field research, points near various sewage outlets on the main and tributary streams of the Lipu Basin were selected to cover different hydrodynamic zones (recharge, runoff, and discharge zones) in the karst basin and river and groundwater samples were collected. Before sampling, the brown glass bottles were rinsed three times with ultrapure water, rinsed three times with sample water, sealed with aluminum foil, and fastened with caps. Parallel samples were set at a ratio of no less than 10% of the total monitoring points for each batch of sampling to ensure accuracy and reliability of the monitoring data. 2 Sample preparation and analysis 2.2.1 Detection of inorganic carbon and nitrogen compounds Immediately after the samples were transported to the laboratory, NH 4 + -N, NO 3 − -N, NO 2 − -N, and inorganic carbon (DIC, including CO 3 2− and HCO 3 − ) were measured. The remaining samples were filtered using a 0.45-µm filter membrane (Tianjin Jinteng, China) and stored in the dark at 0–4°C. Among them, the DIN contents were detected using the JH-TD401 series multi-parameter water quality analyzer based on spectrophotometry principles. DIC was detected using total alkalinity test kits (10–200 mg/L, Lu Heng Bio, China). 2.2.2 Hydrochemical detection and data analysis The anion and cation indicators to be tested included K + , Na + , Ca 2+ , Mg 2+ , SO 4 2− , and Cl − . The samples to be tested were filtered using a 0.22-µm filter membrane and stored away from light at 4°C. Cationic concentrations were determined using an inductively coupled plasma emission spectrometer (Optima 8000, USA) in accordance with the method specified in the National Environmental Protection Standard (HJ 776–2015, China). Anions were detected using ion chromatography (DIONEX ICS-2100, USA). Anions was detected using the method specified in the National Environmental Protection Standard (HJ 84-2016, China). Since the indicator data group consists of multiple measurements of the same object under different conditions and the data are not normal, the Friedman test is used to analyze the significance of the data. 2.3 Isotope sample collection and analysis 2.3.1 Isotope sample collection River water samples were collected onsite using PET sample bottles. First, they were filtered through a 0.45 µm filter membrane to remove various suspended solids and microorganisms. Subsequently, the water samples were stored at temperatures below 4°C. The denitrifying bacteria method was used to determine nitrogen and oxygen isotopes in the samples. The samples to be tested were filtered through a 0.22 µm cartridge filter membrane, and the nitrate concentration was measured before freezing and storage. An appropriate amount of the denitrifying bacterial strain was selected and plated. Afterward, it was incubated at 30°C for 3 days. The monoclonal strain obtained from the culture was obtained and incubated in a test tube containing 6 mL of TSB medium for 8 h in a constant temperature shaker at 32°C and 180 r/min. The remaining strain was stored at 4°C. 2.3.2 Pre-processing and on-machine operation Culture medium (0.2 mL) was drawn and added to 10 mL Tryptic Soy Broth (TSB) culture medium and incubated under the same conditions for 16 h. Finally, 10 mL of the culture medium was transferred to 238 mL of TSB culture medium and incubated under the same conditions for 48 h. After bacterial culture expansion was complete, a colorimetric test was performed. On the next day, 0.2 mL of 10 mol/L NaOH was added with a syringe to terminate the reaction and absorb CO 2 , and then the NO 2 gas generated by the reaction was analyzed on the machine. Hydrogen and oxygen isotopes were measured using a liquid water isotope laser spectrometer. The 13 C stable isotope was commissioned to the Karst Geological Resources and Environment Supervision and Testing Center of the Ministry of Land and Resources (Institute of Karst Geology, Chinese Academy of Geological Sciences, Guilin) for analysis and testing. 2.4 Nitrogen magnification slope and SIAR model Nitrogen Magnification Slope (NMS) is estimated by linear regression of pollutant concentrations and δ 15 N with logarithmic transformation, as shown in Eq. 1 (Lavoie et al. 2013 ): \(\:\text{log}{\text{C}}_{{\text{NO}}_{{\text{3}}^{\text{-}}}}\text{= }\) δ 15 N(b) + a (1) where the slope (b) represents the NMS. A positive slope (TMS > 0) indicates increased N accumulation. A Bayesian stable isotope mixing model was used to quantify the contributions of different nitrate sources to the Lipu River Basin water system. The model was implemented using the SIAR (stable isotope analysis in R) package and can be represented as follows (Ji X et al, 2022): \(\:{\text{X}}_{\text{ij}}\text{=}\sum\:{\text{p}}_{\text{k}}\left({\text{S}}_{\text{ij}}\text{+}{\text{C}}_{\text{ij}}\right)\text{+}{\text{ε}}_{\text{ij}}\) (2) \(\:{\text{S}}_{\text{ij}}\text{=}\text{N}\left({\text{μ}}_{\text{ij}}\text{,}{\text{ω}}_{\text{jk}}^{\text{2}}\right)\) (3) \(\:{\text{C}}_{\text{ij}}\text{=\:}\text{N}\left({{\lambda}^{{\prime\:}}}_{\text{ij}}\text{,}{\text{ω}{\text{τ}}^{{\prime\:}}}_{\text{jk}}^{\text{2}}\right)\) (4) \(\:{\text{ε}}_{\text{ij}}\text{=}\text{N}\left(\text{0,}{\text{σ}}_{\text{j}}^{\text{2}}\right)\) (5) In the above formula (Eq. 1–5), X ij is the JTH isotope value of the i-th mixture, i = 1,2,3,... N; j = 1,2,3,... J; P k is the proportion of the KTH source that needs to be estimated using the SIAR model; S ij is the value of the JTH isotope from the i-th source, following a normal distribution with a mean of µ and a variance of ω; C ij is the fractionation coefficient of the JTH isotope from the i-th source, following a normal distribution with a mean of λ 'and a variance of τ'; ε ij is the residual error, representing the variance that cannot be quantified among the other mixtures, with the mean and variance typically being 0. To estimate the proportional contribution of NO 3 − -N sources, the present study considered two isotopes (j = 2) ( δ 15 N, δ 18 O) and four major potential NO 3 − -N sources (nitrogen fertilizer, soil organic matter, manure, and sewage). Four potential pollution sources were identified: fertilizer, soil, manure, and sewage. The nitrogen and oxygen isotope compositions of the different pollution sources were determined based on previous studies, and the model will be refined further when superior data are obtained. The SIAR model established in the present study was based on R v4.1.3 (R Foundation for Statistical Computing, Vienna, Austria), operated with RStudio 2022.02.0-443, and the code and script were written with reference to the Bayesian mixed model MixSIAR written by Stock. 3 Results and discussion 3.1 Basic nitrogen contents The detection rate of DIN for river water was 100%, with TN concentrations ranging from 0.24 to 16.11 mg/L (average 5.02 ± 4.04 mg/L), which were ( p > 0.05) lower than that in groundwater (6.11 ± 5.01 mg/L). Among them, the average content of the DIN in the river water from high to low was NO 3 − -N (4.27 ± 3.55 mg/L), NH 4 + -N (0.68 ± 0.42 mg/L), NO 2 − -N (0.06 ± 0.04 mg/L), the DIN contents in groundwater, from high to low, were NO 3 − -N (5.75 ± 4.82 mg/L), NH 4 + -N (0.36 ± 0.28 mg/L), NO 2 − -N (0.07 ± 0.05 mg/L). Approximately 15% of the water samples had NO 3 − -N concentrations exceeding the natural threshold for drinking water (10 mg/L; WHO and China). The NH 4 + -N concentration level is significantly ( p < 0.05) higher than that in large rivers with good water quality, such as the middle reaches of the Yangtze River (0.24 mg/L) and the Pearl River (0.33 mg/L) (Zhang et al. 2011 ; Xue et al. 2019 ), and the water quality is roughly the same as that of the main stream of the Yellow River, China, which is greatly affected by local industrial and agricultural activities. Compared with karst river basins abroad, the Lipu River Basin is much better ( p < 0.05) than the Silver Spring in Florida in USA (1.55 mg/L) and the Coesle River Basin in France (2.34 mg/L). In summary, the water quality demonstrates a higher contamination risk compared to major rivers both in China and internationally, reflecting clear evidence of human impact. Figure 2 presents the concentrations of nitrogen in river and groundwater samples from the Lipu River Basin. In Fig. 2 a, the nitrate nitrogen in the river water during summer (average 2.29 ± 1.02 mg/L) was significantly ( p 0.05), indicating that nitrate nitrogen dominates the TN and its seasonal variation is relatively large. 2b showed that the TN of river water in summer (2.95 ± 1.48 mg/L) was significantly ( p < 0.05) lower than that of river water in winter (7.09 ± 6.59 mg/L), while there was no significant difference in the TN samples of groundwater in different seasons. This indicates that karst aquifer systems buffer and homogenize TN, while river water responds more directly and sensitively to seasonal human activities and natural processes. This phenomenon may be related to well-developed faults and folds in the western section of the Nanling Tectonic Belt, where the Lipu Basin is located (Hu, 2024 ), providing a preferred channel for the rapid migration of karst groundwater. 3.1.1 Nitrogen content in river water During the summer, the concentration of NO 3 − -N in the river water ranging from 0.24 to 4.16 (average 2.29 ± 2.01 mg/L), NH 4 + -N (0.64 ± 0.43 mg/L), NO 2 − -N (0.03 ± 0.01 mg/L), with a relative content of 77.57, 21.68 and 0.75%. NO 3 − -N concentrations ranged from 0.24 to 4.16 mg/L (6.25 ± 5.65 mg/L), NH 4 + -N (0.76 ± 0.65 mg/L), NO 2 − -N (0.09 ± 0.05 mg/L), with a relative content of 88.03, 10.70 and 1.27%. Overall, the relative content of NO 3 − -N was more than 77%, indicating that the nitrogen form entering the river was mainly NO 3 − -N, and the sources were mainly nitrified or rapidly nitrified pollution sources, including fertilizer application, livestock breeding, and sewage treatment plant discharge in the study area (Yu Z et al. 2022). The average concentration of NO 3 ⁻-N in winter (6.25 mg/L) was significantly higher ( p < 0.05) than that in summer (2.29 mg/L), and the fluctuation range was larger. Meanwhile, NH 4 + -N and NO 3 ⁻-N also saw a slight increase in winter. This indicates that water bodies in winter may be affected by factors such as stronger external nitrogen input or weakened absorption by aquatic plants in summer (Friedrichs et al. 2019 ). The nitrogen concentration is relatively high near industrial land, with NO 3 − -N (6.65 ± 5.63 mg/L), NH 4 + -N (0.75 ± 0.53 mg/L), NO 2 − -N (0.04 ± 0.03 mg/L). However, there was no significant ( p > 0.05) difference in DIN concentration under industrial, agricultural and other land use types, indicating that the pollutants in the river water underwent intense mixing and homogenization. Speculative study area had a mixed connection and discharge in the river sewage collection system (Shoaib et al. 2008 ), which led to further mixing of pollutants during transportation. 3.1.2 Nitrogen content in groundwater The TN concentration of groundwater in the basin (average 6.12 ± 5.33 mg/L) was higher than that of river water, indicating persistent pollution seepage and accumulation. Among them, the NH 4 + -N concentration in groundwater during the summer (0.34 ± 0.31 mg/L) with a relative content of TN was 5.25%. NO 3 − -N concentration (5.98 ± 5.11 mg/L). NO 2 − -N concentration (0.02 ± 0.01 mg/L) account for 0.22% of the TN content. The concentration of NH 4 + -N in groundwater during the winter (0.37 ± 0.34 mg/L) with a relative content of TN was 6.44%. NO 3 − -N concentration (5.36 ± 5.21 mg/L) NO 2 − -N concentrations (0.03 ± 0.03 mg/L) account for 0.22% of the TN content. The concentration of NO 2 − -N in the summer was significantly ( p < 0.05) higher than in the winter, with increased rainfall in the summer and river runoff scouring dissolved NO 3 − -N precursors into the vesicular zone for infiltration. In karst basins, rapid infiltration through fissure-dominated flow pathways facilitates pollutant transport to phreatic aquifers during summer, leading to elevated NO 2 ⁻-N concentrations (Chu et al. 2021 ). The NO 3 − -N concentration (4.53 ± 3.88 mg/L) with the annual average concentration in the Central High-Tech Industrial Park exceeding the natural threshold for drinking water (10 mg/L; WHO and China). IW discharge outlets include industries such as chemical, pharmaceutical, and electroplating industries, which use large amounts of raw materials such as nitrates and nitrites in production processes (Huo et al. 2019 ; Li et al. 2025 ). The nutrients may partially remain in the wastewater, in turn increasing the concentrations of NO 3 − -N in the wastewater. The average NH 4 + -N in groundwater was 0.41 mg/L, which was slightly lower than the concentration in river water bodies, and the pH of water in karst basins was weakly alkaline (pH 7.5–8.5). Studies on the pre-loss of NH 4 + -N volatilization in rivers under alkaline conditions and the specificity of NH 4 + -N volatilization in karst areas suggest that the amount of NH 4 + -N entering groundwater is relatively low (Woodward et al. 2011 ). Unlike in river water, the NO 3 − -N concentration in industrial wastewater was significantly different from those in sanitary wastewater and other types, indicating that the hydrological paths of groundwater and river water were different. The main source of NO 3 − -N pollution in groundwater in the basin is closely related to hanger manufacturing and specialty food processing industries that are dominant by Lipu County (M. Zhang et al. 2023 ). Nitrogen sources are converted to NO 3 − -N through NH 3 oxidation and nitrification during wastewater treatment, resulting in an increase in discharge concentration, conversion of wastewater with high organic N from food processing, and natural oxidation of nitrite, eventually generating NO 3 − -N into groundwater. This finding aligns with Wu’s probabilistic assessments (Wu et al. 2025 ). Groundwater discharge data suggest that the lack of significant differences in N 3 in river water does not deny the differences in the characteristics of the discharge sources themselves, but rather highlights the fundamental differences in pollution plume mixing capacity between the river river water environment and the underground aquifer environment in the region. 3.2 Spatial distribution 3.2.1 Nitrogen content in river water During the observation period, the concentration of TN in the river water (average 4.95 ± 4.36 mg/L). Taking the highest proportion of nitric nitrogen as the benchmark, according to the WHO and China standards, the proportion of river monitoring points exceeding the threshold is 33.95%. The average concentrations in the tributaries of the Puling, Dumo, Datang, and Limu rivers in the basin were relatively high, and the lower reaches of the Maling River points were affected by the grain products and manufacturing processing industrial park, and the concentration of N was relatively high, increasing by 2.2 mg/L compared with that upstream of the tributaries. The concentrations of various forms of nitrogen in the main stream are NH 4 + -N (0.55 ± 0.38 mg/L), NO 3 − -N (2.07 ± 1.78 mg/L), and NO 3 − -N (0.02 ± 0.01 mg/L). In the tributaries, they are respectively NH 4 + -N (2.21 ± 1.87 mg/L), NO 3 − -N (2.52 ± 2.38 mg/L), and NO 2 − -N (0.01 ± 0.01 mg/L). In the main stream, the Lipu River showed a low trend in the middle reaches and a high trend in the upper and lower reaches. The main reasons are that the middle reaches fall within the urban area of Lipu County, where there are many hospitals and municipal sewage outlets, and the industrial types near DT2 are complex, covering metal processing, oil industry, rice, and flour products, etc. Strong karst development in the tributaries leads to low DO (< 2 mg/L), and an anoxic environment severely inhibits nitrification (Diab et al. 1993 ), resulting in the accumulation of NH 4 + -N in the tributaries. The TN pollution levels in tributaries were higher ( p < 0.05) than those in the main stream (5.11 mg/L, 2.64 mg/L), and the flow rates in tributaries were mostly over 3.5 m 3 /s, which is more than 30% of the flow rate of the main stream. The pollution load is relatively high, indicating that tributaries such as the Pulu and Dumo Rivers are the core sources of N pollution in the basin (Fakouri et al. 2022 ). The river water monitoring points were mainly driven by OT and SW. Points L5, DM1, and DM2 were affected by nearby karst skylights, with higher DO (5 mg/L) and NO 3 − -N concentrations of 3.05, 4.16, and 2.93 mg/L, respectively. Near PL2, L6, and DT2, there were large karst springs with NO 3 − -N concentrations of 3.11, 2.97 and 3.05 mg/L respectively, all higher than the average NO 3 − -N concentration (2.88 mg/L) in the basin, confirming the earlier conclusion of groundwater recharge to the river water. The monitoring data near the karst springs and skylights were significantly different from those at other points in the basin ( p < 0.05). Considering the regional monitoring points were pure carbonate rock hydrogeological conditions, the retention time of pollutants around the large springs and skylights was potentially shortened (Meng et al. 2021 ) and the natural degradation rate was relatively low. 3.2.2 Nitrate-nitrogen content in groundwater The average groundwater TN concentration was 5.743 mg/ L. Based on the WHO and China standards, 18.83% of groundwater monitoring points were exceeding the threshold. Affected by industrial parks, such as metalworking, the concentration of TN at monitoring points near U10 and 11 was 2–3 times higher than the overall regional concentration (Fig. 5cd)). This indicates that the IW outlet is a "hot source area" for triazine, which directly leads to a sudden increase in triazine concentration in the surrounding wells. The upstream agricultural area has become a significant source of groundwater NO 3 − -N pollution, and the fundamental driving factor is extensive application of organic fertilizer in the Lipu Taro planting model and the subsequent nitrification process. The concentrations of NO 3 − -N in the U1–U3 region were 1.65–3.57 mg/L higher than the average concentration of groundwater in the region. Organic fertilizers can account for 30%–50% of the cultivation of Lipu taro in the region. Organic nitrogen is converted into NH 4 + -N through microbial mineralization and then into NO 3 − -N by nitrifying bacteria. If the application is excessive or uncomposted, the rate of nitrate formation exceeds the absorption capacity of the crop (Cui et al. 2023 b), resulting in accumulation of NO 3 − -N. U4 and U5 were affected by nearby karst skylights, with NO 3 − -N concentrations of 2.87 and 3.2 mg/L, respectively. There are large karst springs near U10, U11, and U12, with NO 3 − -N concentrations of 6.1, 18.3, and 5.6 mg/L, respectively. The monitoring data near the karst spring showed no significant differences from the other points in the basin; however, the monitoring points near the karst skylight showed significant differences from those of the other points (P < 0.05). The hydrogeological types at points U10, U11, and U12 were bedrock fissure water, which was restricted by atmospheric precipitation recharge, with short underground runoff and scarce water volume (Yi et al. 2023 ). The flow rate of the spring was generally 0.014–0.325 L/s, and the average concentration of NO 3 − -N at U10, U11, and U12 was relatively high (10 mg/L), which was also affected by discharge from the new materials industrial park. 3.3 Seasonal variation River water TN (average 2.96 ± 2.36 mg/L) groundwater TN (5.77 ± 4.86 mg/L) during the winter (Fig. 5 , 6 ). In river water, the concentrations of NO 3 − -N, NH 4 + -N, and NO 2 − -N from October to March of the following year were higher than those from July to September, mainly because of abundant precipitation during the summer and enhanced river runoff, which diluted the three N contents in the water body. During the winter, there is less precipitation and water flow slows down, and triazine accumulation in the basin leads to higher concentration. Sewage outlets within the basin also affect the distribution of nitrogen; however, the dilution and accumulation effects caused by seasonal changes are more significant. Notably, there was a significant difference in NO 2 − -N between the wet and winters, with concentrations in the summer being approximately three times those in the winter. The unique hydrological and water quality conditions in the study area during the summer (high load input, turbidity, and anoxic microenvironment) disrupted N transformation (Zhao et al. 2020 ) and inhibited the nitrite consumption pathway (Fig. 5 b). In groundwater, the high value areas of the DIN are often concentrated at the IW monitoring point (Fig. 5cd), which reflects both the continuous influence of industrial pollution sources and the fissure-pipe type flow characteristics of karst basins. Once pollutants enter a karst aquifer through infiltration, they migrate to relatively connected channels, forming relatively stable high-value areas. The proportion of nitrogen in river water changed during the wet and winters (the proportion of NO 3 − -N decreased from 88.5% to 70.2%) (Fig. 5 ). During the summer, NO 3 − -N-dominated nonpoint source pollution is prevalent, and the hydrochemical characteristics are strongly influenced by agricultural nonpoint sources, whereas the point source pollution characteristics are masked. The winter is dominated by point source pollution (with an increase of 10–20 times in NO 3 − -N). The water chemical characteristics reflect the direct influence of OT and IW, with a prominent contribution of point source pollution. NO 3 − -N, such as in DT1, LM2, and L13, was altered from a secondary component during the summer (a) to a dominant component among the DIN sources. Groundwater did not exhibit such a trend, but the proportion of NO 2 − -N at some monitoring points during the winter increased 2–3 times compared with that in the summer. The conclusion can be explained by the denitrification process, in which the karst aquifer is in an anoxic microenvironment, which inhibits further conversion of NO 2 − -N to NO 3 − -N. 3.4 Relationship between NO 3 − and each water quality index 3.4.1 NO 3 − and isotope There was a positive correlation between the logarithmic conversion concentration of NO 2 − -N and δ 15 N in SW and IW (Fig. 6 ), and NO 2 − -N from manure/sewage sources in industrial and agricultural areas conformed to the characteristics of high NO 2 − -N accompanied by high δ 15 N (Serna et al. 2010 ), at which point δ 15 N is higher (10 to 25‰). In contrast, an inverse relationship was observed between OT logarithmic conversion concentration and δ 15 N, and the dominant factor was presumed to be the very low δ¹⁵N value of synthetic fertilizers (such as urea, ammonium salts, nitrate fertilizers), close to the atmospheric δ 15 N value (0‰), generally ranging from − 6‰ to + 6‰. When fertilizers are applied in large quantities and not fully absorbed by plants or fixed in soil, the concentration of NO 2 − -N entering the discharge outlet through runoff or leaching is very high (Sturm et al. 2011), whereas the δ 15 N value is relatively low. The logarithmic conversion concentrations of NO 3 − at SW and IW monitoring points were positively correlated with δ 18 O, suggesting the presence of different sources of nitrate (domestic, industrial, soil mineralization) in the water body, with significant differences in δ 18 O values from these sources (Brooks et al. 2012 ), and positive correlation between NO 2 − -N concentration and δ 18 O may occur after mixing. The nitrifications at SW and IW monitoring points were stronger than that at OT, the NO 2 − -N concentration increased, and the δ 18 O value approached the δ 18 O value of the water source, so that it was positively correlated with the NO 2 − -N concentration. 3.4.2 NO 3 − and ion The degree of mineralization was determined based on electrical conductivity (EC) values using Cl − conserved tracer sources (Siddique et al. 2016 ). The relationship between Cl − and NO 3 − , and between EC and NO 3 − , can determine whether the change in NO 3 − concentration in water can be attributed to the mixed process of N-free transformation or denitrification. If the increase in nitrate concentration is accompanied by increases in Cl − and EC concentrations, the NO 3 − concentration in the water may be affected by the mixed process of nitrogen form transformation. That is, denitrification is the main nitrogen transformation process. The EC and Cl − in the water bodies of all types of sewage outlets were positively correlated (Fig. 7 ), and the correlation was the highest at industrial sewage outlets (R 2 = 0.596), indicating that wastewater from industrial parks in the study area often contained chlorinated compounds (such as hydrochloric acid and sodium chloride). High-temperature wastewater is produced during sterilization, disinfection, cooking, and drying (Bartkowiak et al. 2020 ). When the temperature increased, the mobility of the chloride ions increased, as did the EC. NO 3 − and Cl − were positively correlated at all discharge outlets, which is a relationship that is prevalent in various pollution scenarios and reflects the pattern of synergistic changes among pollutants. The data show that the linear relationship between NO 3 − and Cl − concentrations in groundwater in the basin is not significant. In addition to anthropogenic pollution sources, this may be due to karst inhomogeneity in the vertical direction, with some monitoring points under hypoxic or anoxic conditions (Lorette et al. 2022 ), resulting in local denitrification. 3.5 Source analysis of isotopes based on SIAR model The SIAR model was used to calculate the contribution rates of the five pollution sources to NO 3 − -N in the river water and groundwater in the study area. The largest source of pollution in the basin's river water was M&S, followed by SN, possibly because the river water was affected considerably by river runoff (Cui et al. 2020 ), and heavy rainfall during the summer washed the homestead soil, carrying high amounts of organic nitrogen into the river water. Studies have shown that some of the organic nitrogen is utilized by nitrification after being introduced into river water and is not transported to groundwater (Xiong et al. 2023 ), but soil remains the second largest source of nitrogen in groundwater, which may be either transported from river water or directly released into groundwater by soil leaching. The largest sources of nitrogen in groundwater are manure and sewage. Garbage stations may use sewage to flush landfills that do not produce much runoff, most of the water will seep directly, and the process will also leach the soil, which explains the close ratio of soil to sewage sources in the industrial emissions areas (Vadas et al. 2008 ). The δ 18 O-NO 3 − values of the water samples in the study area were all within the typical δ 18 O-NO 3 − range produced by nitrification (Li et al. 2019 ), indicating that nitrification is a key process for establishment of the source of NO 3 − in river water and groundwater. In groundwater dominated by nitrification, the higher the dissolved oxygen content, the higher the δ 15 N-NO 3 − value. The nitrogen and oxygen isotope fractionation produced during denitrification followed Rayleigh fractionation, and the δ 15 N and δ 18 O values of the remaining NO 3 − satisfied the Rayleigh formula, showing that as NO 3 − concentration decreased, both δ 15 N and δ 18 O values increased and were linearly correlated. In addition, the slope of the enrichment coefficient ratio (ε N /ε O ) varied between 1.3 and 2.1. Most of the data at the monitoring points were concentrated in the left zone of ε N /ε O = 2.1, where the NO 3 − -N concentration and the values of ni and oxygen isotopes were within a certain range, indicating that the changes in the water body were not significant and there was no obvious denitrification. The δ 15 N and δ 18 O values in groundwater were significantly inversely proportional to the NO 3 − -N concentration, and the slope of the enrichment coefficient ratio of the two was about 1.38, suggesting denitrification in groundwater (Yuan et al. 2020 ), combined with the fact that most of the δ 15 N and δ 18 O values in groundwater were low. The analyzed water samples exhibited suboxic conditions (DO < 2 mg/L), indicating that denitrification was not intense. The δ 15 N and δ 18 O values in groundwater were slightly higher than those in river water, suggesting that groundwater was more strongly influenced by fecal matter and domestic sewage inputs. This is likely attributable to the widespread use of dry toilet systems in rural areas and the application of manure to agricultural land (Wang et al. 2021 ). The δ 18 O-NO 3 − of agricultural area was higher than that of other areas, suggesting that major nitrogen biogeochemical processes have changed across different pollution types. Many studies have found that pollution type and water body characteristics largely influence the NO 3 − isotopic characteristics (Choi et al. 2020 ; Yu et al. 2022 ). The δ 15 N-NO 3 − values observed in IA were lower than those observed in FA, mainly because SN and NF are the main sources of NO 3 − (Kim et al. 1996 ). The enrichment of NO 3 − isotopes in river water may be due to the increased consumption of nitrogen fertilizer and intensified volatilization of livestock manure during the period, as well as the evaporation of leachate from landfills, septic tanks, and domestic waste at high temperatures. However, δ 18 O-NO 3 − values are influenced mainly by microbial nitrification, and not by nitrogen sources, land use, or seasonal variations. Overall, the higher δ 15 N-NO 3 − in groundwater is mainly attributed to M&S discharge, as well as higher population density and garbage leachate. 4 Conclusion This study establishes a novel probabilistic framework for nitrate source apportionment in karst aquifer systems by integrating dual stable isotope analysis (δ 15 N-NO 3 − and δ 18 O-NO 3 − ) with Bayesian mixing models (SIAR) and multivariate hydrochemical indicators. The key findings reveal distinct stochastic patterns in nitrogen dynamics: Groundwater TN concentrations (6.12 ± 5.33 mg/L) systematically exceeded river water levels (5.02 ± 4.04 mg/L), with seasonal variability amplified by karst-specific rapid recharge pathways. Spatially, tributaries adjacent to industrial zones exhibited elevated nitrogen levels, particularly near the grain processing and metal manufacturing facilities. Hydrochemical and isotopic evidence revealed source-specific signatures. Significant positive correlations between NO₃⁻ and both Cl⁻ and EC (R² = 0.596) in industrial wastewater indicated high dissolved salt co-discharge with limited in-stream denitrification. The winter correlation between NO 3 ⁻ and SO 4 2 ⁻ confirmed co-application of sulfur- and nitrogen-containing fertilizers in agricultural areas. The δ 15 N-δ 18 O enrichment slope of 1.38, combined with δ 18 O-NO 3 ⁻ values ranging from − 10‰ to 10‰, demonstrates that nitrification dominates over denitrification in the basin's groundwater system—a critical finding for understanding nitrogen fate in karst aquifers. SIAR modeling quantified manure and sewage as the dominant nitrate source, contributing 56.7 ± 8.2% of total inputs, demonstrating the statistical robustness achievable through Bayesian inference in complex multi-source environments. Based on these findings, we recommend: (1) Implement an integrated source-pathway-receptor protection strategy strategic livestock relocation from karst-sensitive zones to consolidated feeding operations with advanced manure treatment systems (targeting 30% reduction in diffuse M&S sources within 5 years) and (2) given the accumulation of nitrogen load during the winter months, it is essential to collaborate with local agricultural cooperatives and food processing industries—particularly those involved in taro and grain production—to implement targeted nitrogen monitoring programs specifically for the winter season. This integrated isotopic-hydrochemical framework offers a transferable methodology for nitrogen source apportionment in heterogeneous karst systems. Declarations Acknowledgments This work was supported by the Guangxi University Engineering Research Center for River Basin Protection and Green Development, Guilin University of Technology and the Guangxi Engineering Research Center for Comprehensive Control of Agricultural Non-point Source Pollution. Funding This work was funded by the National Natural Science Foundation of China (grant 52360001) and the Guangxi Basic Ability Enhancement Programme for Young and Middle-Aged Teachers (grant 2023KY0273). Author contributions Zupeng Wan: Conceptualization, Data curation, Form analysis, model, Writting —Original draft, Writing—reviewing and editing, and Funding acquisition. Baojun Liu: Data curation, Investigation. Xiaoyu Yan: Form analysis. Yingjie Chen: Conceptualization. Wenwen Chen: Resources, Methodology. Honghu Zeng * : Writing—reviewing and editing, Funding acquisition. Competing interests The authors declare no competing interests. References Bartkowiak A, Bartkowiak A, Jaworska H, Dąbkowska-Naskręt H, Rydlewska M (2020) Effect of salinity on the mobility of trace metals in soils near a soda chemical factory. J Elem 25:501–512. https://doi.org/10.5601/jelem.2019.24.2.1875 Biddau R, Dore E, Da Pelo S, Lorrai M, Botti P, Testa M, Cidu R (2023) Geochemistry, stable isotopes and statistic tools to estimate threshold and source of nitrate in groundwater (Sardinia, Italy). Water Res 232:119663. https://doi.org/10.1016/j.watres.2023.119663 Boumaiza L, Walter J, Chesnaux R, Huneau F, Garel É, Erostate M, Johannesson KH, Vystavna Y, Bougherira N, Bordeleau G, Stotler RL, Blarasin M, Gutiérrez M, Knöller K, Stumpp C (2022) Multi-tracer approach to understand nitrate contamination and groundwater-surface water interactions in the Mediterranean coastal area of guerbes-senhadja, algeria. J Contam Hydrol 251:104098. https://doi.org/10.1016/j.jconhyd.2022.104098 Brooks JR, Wigington PJ, Phillips DL, Comeleo R, Coulombe R (2012) Willamette River Basin surface water isoscape (δ 18 O and δ 2 H): temporal changes of source water within the river. Ecosphere 3:1–21. https://doi.org/10.1890/ES11-00338.1 Chang Y, Qi Y, Wang Z (2024) Comprehensive evaluation of IMERG, ERA5-Land and their fusion products in the hydrological simulation of three karst catchments in Southwest China. J Hydrol Reg Stud 52:101671. https://doi.org/10.1016/j.ejrh.2024.101671 Choi J-Y, Kim S-K, Kim J-C, La G-H (2020) Utilization of nitrate stable isotopes of Chydorus sphaericus (OF Müller) to elucidate the hydrological characteristics of riverine wetlands in the Nakdong River, South Korea. J Ecol Environ 44:1. https://doi.org/10.1186/s41610-019-0144-1 Chu X, Ding H, Zhang X (2021) Simulation of solute transport behaviors in saturated karst aquifer system. Sci Rep 11:15614. https://doi.org/10.1038/s41598-021-94950-7 Cui N, Cai M, Zhang X, Abdelhafez AA, Zhou L, Sun H, Chen G, Zou G, Zhou S (2020) Runoff loss of nitrogen and phosphorus from a rice paddy field in the east of China: Effects of long-term chemical N fertilizer and organic manure applications. Glob Ecol Conserv 22:e01011. https://doi.org/10.1016/j.gecco.2020.e01011 Cui R, Zhang D, Hu W, Zhao X, Yan H, Liu G, Chen A (2023) Nitrogen in soil, manure and sewage has become a major challenge in controlling nitrate pollution in groundwater around plateau lakes, Southwest China. J Hydrol 620:129541. https://doi.org/10.1016/j.jhydrol.2023.129541 Cui R, Zhang D, Hu W, Zhao X, Yan H, Liu G, Chen A (2023) Nitrogen in soil, manure and sewage has become a major challenge in controlling nitrate pollution in groundwater around plateau lakes, Southwest China. J Hydrol 620:129541. https://doi.org/10.1016/j.jhydrol.2023.129541 Diab S, Kochba M, Avnimelech Y (1993) Nitrification pattern in a fluctuating anaerobic-aerobic pond environment. Water Res 27:1469–1475. https://doi.org/10.1016/0043-1354(93)90027-F Fakouri B, Mohamad Vali Samani J, Mohamad Vali Samani H, Mazaheri M (2022) Cost-based model for optimal waste-load allocation and pollution loading losses in river system: simulation–optimization approach. Int J Environ Sci Technol 19:12103–12118. https://doi.org/10.1007/s13762-022-04422-2 Friedrichs MAM, St-Laurent P, Xiao Y, Hofmann E, Hyde K, Mannino A, Najjar RG, Narváez DA, Signorini SR, Tian H, Wilkin J, Yao Y, Xue J (2019) Ocean circulation causes strong variability in the mid-atlantic bight nitrogen budget. J Geophys Res Oceans 124:113–134. https://doi.org/10.1029/2018JC014424 Grechanik SV, Klymenko NA, Bunetskyi VA, Smolin SK, Zabneva OV, Nevynna LV (2024) Production of Activated Biochar from Wood Raw Materials for Water Treatment and Water Purification Applications. J Water Chem Technol 46:512–523. https://doi.org/10.3103/S1063455X24050047 Guo W, Zhang K, Liang Z, Zou R, Xu Q (2019) Electrochemical nitrogen fixation and utilization: theories, advanced catalyst materials and system design. Chem Soc Rev 48:5658–5716. https://doi.org/10.1039/C9CS00159J Güven D, Hanhan O, Aksoy EC, Insel G, Çokgör E (2017) Impact of paint shop decanter effluents on biological treatability of automotive industry wastewater. J Hazard Mater 330:61–67. https://doi.org/10.1016/j.jhazmat.2017.01.048 Huo J, He G, Chen W, Hu X, Deng Q, Chen D (2019) A minireview of hydroamination catalysis: alkene and alkyne substrate selective, metal complex design. BMC Chem 13:89. https://doi.org/10.1186/s13065-019-0606-7 Iqbal J, Su C, Wang M, Abbas H, Baloch MYJ, Ghani J, Ullah Z, Huq ME (2022) Groundwater fluoride and nitrate contamination and associated human health risk assessment in South Punjab. Pakistan Environ Sci Pollut Res 30:61606–61625. https://doi.org/10.1007/s11356-023-25958-x Iriawan H, Andersen SZ, Zhang X, Comer BM, Barrio J, Chen P, Medford AJ, Stephens IEL, Chorkendorff I, Shao-Horn Y (2021) Methods for nitrogen activation by reduction and oxidation. Nat Rev Methods Primer 1:56. https://doi.org/10.1038/s43586-021-00053-y Ji X, Shu L, Chen W, Chen Z, Shang X, Yang Y, Dahlgren RA, Zhang M (2022) Nitrate pollution source apportionment, uncertainty and sensitivity analysis across a rural-urban river network based on δ 15 N/δ 18 O-NO 3 – isotopes and SIAR modeling. J Hazard Mater 438:129480. https://doi.org/10.1016/j.jhazmat.2022.129480 Kim CS, Sandretto C, Hostetler J (1996) Effects of Farmer Response to Nitrogen Fertilizer Management Practices on Groundwater Quality. Water Resour Res 32:1411–1415. https://doi.org/10.1029/96WR00491 Lai Y, Yang H, Qiu F, Dang Z, Luo Y (2023) Can Rural Industrial Integration Alleviate Agricultural Non-Point Source Pollution? Evidence from Rural China. Agriculture 13:1389. https://doi.org/10.3390/agriculture13071389 Lavoie RA, Jardine TD, Chumchal MM, Kidd KA, Campbell LM (2013) Biomagnification of Mercury in Aquatic Food Webs: A Worldwide Meta-Analysis. Environ Sci Technol 47:13385–13394. https://doi.org/10.1021/es403103t Li H, Tian R, Zhang Y (2025) Electrochemically superior ammonia production from electroplating wastewater over oxygen-incorporated antiperovskite. J Colloid Interface Sci 683:49–57. https://doi.org/10.1016/j.jcis.2024.12.122 Li X, Tang C, Cao Y, Li X (2019) Carbon, nitrogen and sulfur isotopic features and the associated geochemical processes in a coastal aquifer system of the Pearl River Delta, China. J Hydrol 575:986–998. https://doi.org/10.1016/j.jhydrol.2019.05.092 Li Y, Yang J, Wen J (2023) Entropy-based redundancy analysis and information screening. Digit Commun Netw 9:1061–1069. https://doi.org/10.1016/j.dcan.2021.12.001 Lin H, Yan S, Song B, Wang Z, Sun L (2019) Discrimination of aged rice using colorimetric sensor array combined with volatile organic compounds. J Food Process Eng 42:e13037. https://doi.org/10.1111/jfpe.13037 Lorette G, Sebilo M, Buquet D, Lastennet R, Denis A, Peyraube N, Charriere V, Studer J-C (2022) Tracing sources and fate of nitrate in multilayered karstic hydrogeological catchments using natural stable isotopic composition (δ 15 N-NO 3 – and δ 18 O-NO 3 – ). Application to the Toulon karst system (Dordogne, France). J Hydrol 610:127972. https://doi.org/10.1016/j.jhydrol.2022.127972 Ma Y, Jones DL, Wang J, Cardenas LM, Chadwick DR (2021) Relative efficacy and stability of biological and synthetic nitrification inhibitors in a highly nitrifying soil: Evidence of apparent nitrification inhibition by linoleic acid and linolenic acid. Eur J Soil Sci 72:2356–2371. https://doi.org/10.1111/ejss.13096 Meng Q, Xing L, Liu L, Xing X, Zhao Z, Zhang F, Li C (2021) Time-lag characteristics of the response of karst springs to precipitation in the northern China. Environ Earth Sci 80:348. https://doi.org/10.1007/s12665-021-09640-4 Pandit S, Grassian VH (2022) Gas-phase nitrous acid (HONO) is controlled by surface interactions of adsorbed nitrite on common indoor material surfaces. Environ Sci Technol 56:12045–12054. https://doi.org/10.1021/acs.est.2c02042 Schaefer MV, Ying SC, Benner SG, Duan Y, Wang Y, Fendorf S (2016) Aquifer Arsenic Cycling Induced by Seasonal Hydrologic Changes within the Yangtze River Basin. Environ Sci Technol 50:3521–3529. https://doi.org/10.1021/acs.est.5b04986 Serna A, Pätsch J, Dähnke K, Wiesner MG, Christian Hass H, Zeiler M, Hebbeln D, Emeis K-C (2010) History of anthropogenic nitrogen input to the German Bight/SE North Sea as reflected by nitrogen isotopes in surface sediments, sediment cores and hindcast models. Cont Shelf Res 30:1626–1638. https://doi.org/10.1016/j.csr.2010.06.010 Shoaib AM, Aly SM, Awad ME, Foo DCY, El-Halwagi MM (2008) A hierarchical approach for the synthesis of batch water network. Comput Chem Eng 32:530–539. https://doi.org/10.1016/j.compchemeng.2007.03.015 Siddique AA, Dixit MK, Tembe BL (2016) Molecular dynamics simulations of Ca 2+ Cl – ion pair in polar mixtures of acetone and water: Solvation and dynamical studies. Chem Phys Lett 662:306–316. https://doi.org/10.1016/j.cplett.2016.09.061 Šturm M, Kacjan-Maršić N, Lojen S (2011) Can δ 15 N in lettuce tissues reveal the use of synthetic nitrogen fertiliser in organic production? J Sci Food Agric 91:262–267. https://doi.org/10.1002/jsfa.4179 Vadas PA, Owens LB, Sharpley AN (2008) An empirical model for dissolved phosphorus in runoff from surface-applied fertilizers. Agric Ecosyst Environ 127:59–65. https://doi.org/10.1016/j.agee.2008.03.001 Wang S, Lin W, Ye Q, Lv W, Liao P, Yu J, Mu C, Wu L, Muneer MA, Zhang Y, Zhan R, Wu A, Wu H, Chen C, Su H, Chen Y, Wang Z, Zhang Z, Ji H, Guo L (2024) Effects of Different Nitrogen Fertilizer Rates on Soil Magnesium Leaching in Tea Garden. J. Soil Sci. Plant Nutr 24:6630–6640. https://doi.org/10.1007/s42729-024-01995-4 Wells NS, Maher DT, Erler DV, Hipsey M, Rosentreter JA, Eyre B.D, n.d. Estuaries as Sources and Sinks of N 2 O Across a Land Use Gradient in Subtropical Australia. Glob. Biogeochem. Cycles Woodward TR, Frame WH, Alley MM, Whitehurst GB, Whitehurst BM (2011) Design and validation of a laboratory system for measurement of volatilized ammonia. Agron J 103:38–44. https://doi.org/10.2134/agronj2010.0242 Wu H, Wang Y, Liu Y, Guo S, Zhong Z (2025) Experimental and simulation study of flue gas desulfurization using the limestone-gypsum wet method under an oxygen-enriched combustion atmosphere. J Energy Inst 119:101999. https://doi.org/10.1016/j.joei.2025.101999 Wu P, Xiao Q, Guo Y, Prelovšek M, Yu Q, Wang Q (2022) Migration, transformation and nitrate source in the Lihu Underground River based on dual stable isotopes of δ 15 N-NO 3 – and δ 18 O-NO 3 – . Environ Sci Pollut Res 29:48661–48674. https://doi.org/10.1007/s11356-022-19277-w Xiong G, Zhu X, Liu M, Yang Y, Chen G, Fu T, Ding R, Xu X, Wu J (2023) Nitrogen cycle pattern variations during seawater-groundwater-river interactions enhance the nitrogen availability in the coastal earth critical zone. J Hydrol 624:129932. https://doi.org/10.1016/j.jhydrol.2023.129932 Yan J, Chen J, Zhang W (2023) A new probabilistic assessment process for human health risk (HHR) in groundwater with extensive fluoride and nitrate optimized by non parametric estimation method. Water Res 243:120379. https://doi.org/10.1016/j.watres.2023.120379 Yang Y, Wu J, Du J, Yang, Yunhua, Ye F, Wang Y, Hong Y (2024) Coupled Stable Isotope Tracing and Sulfamic Acid Reduction (SIT–SAR) Method to Determine the Ammonia and Nitrite Oxidation Rates in Water and Sediments. Anal Chem 96:4197–4204. https://doi.org/10.1021/acs.analchem.3c05461 Yi L, Xia R, Tang J, Lu D, Wang Z, Lu H, Zhao L (2023) Modeling the effect of the submerged zone on nitrogen removal efficiency of a bioretention system under dry-wet alterations. J Hydrol 623:129788. https://doi.org/10.1016/j.jhydrol.2023.129788 Yi L, Xia R, Tang J, Lu D, Wang Z, Lu H, Zhao L (2015) Karst conduit hydro-gradient nonlinear variation feature study: case study of Zhaidi karst underground river. Environ Earth Sci 74:1071–1078. https://doi.org/10.1007/s12665-014-3963-5 Yu L, Zheng T, Yuan R, Zheng X, Liu P, Zhang C, Xue C, Mu Y, Liu J, Zhang Y, Tian D, Ye C, Zhang H, Guan J, Iqbal J, Su C, Wang M, Abbas H, Baloch MYJ, Ghani J, Ullah Z, Huq ME (2022) APCS-MLR model: A convenient and fast method for quantitative identification of nitrate pollution sources in groundwater. J Environ Manage 314:115101. https://doi.org/10.1016/j.jenvman.2022.115101 Yu Z-Q, Nakagawa K, Berndtsson R, Hiraoka T, Suzuki Y (2022) Effects of the Japanese Nitrate Directive Plan (NDP) to curb groundwater nitrate-nitrogen content in the Miyakonojo River basin. J Hydrol 615:128563. https://doi.org/10.1016/j.jhydrol.2022.128563 Yuan Z, Liao Y, Zheng M, Zhuo M, Huang B, Nie X, Wu X, Li D (2020) Relationships of nitrogen losses, phosphorus losses, and sediment under simulated rainfall conditions. J Soil Water Conserv 75:231–241. https://doi.org/10.2489/jswc.75.2.231 Zhang C, Luo X, Deng Y, Deng Z, Xu R, Amer MA, Ali EAE, Jiang J, Chen H (2024) Insights into enhanced pollutant removal from road runoff by functional microorganisms in a field-scale bioretention facility. J Water Process Eng 62:105294. https://doi.org/10.1016/j.jwpe.2024.105294 Zhang M, Forte-Perri V, Sun W, Tang L, Huang S, Guo T, Chen X, Li Q (2023) Identification and Observation of Infection Processes of Colletotrichum Species Associated with Persimmon Anthracnose in Guangxi, China. J Geochem Explor 107:1670–1679. https://doi.org/10.1094/PDIS-06-22-1372-SR Zhang P, Wang X-D, Yue F-J, Chen S-N, Liu Z-H, Lu M, Shi Z-Y (2023) Dynamic characteristics of nitrogen transport in various land use in a typical karst catchment during rainfall events. Sci Data 82:332–343. https://doi.org/10.1007/s12665-023-10980-6 Zhang P, Wang X-D, Yue F-J, Chen S-N, Liu Z-H, Lu M, Shi Z-Y, Olarinoye T, Gleeson T, Marx V, Seeger S, Adinehvand R, Allocca V, Andreo B, Apaéstegui J, Apolit C, Arfib B, Auler A, Bailly-Comte V, Barberá JA, Batiot-Guilhe C, Bechtel T, Binet S, Bittner D, Blatnik M, Bolger T, Brunet P, Charlier J-B, Chen Z, Chiogna G, Coxon G, De Vita P, Doummar J, Epting J, Fleury P, Fournier M, Goldscheider N, Gunn J, Guo F, Guyot JL, Howden N, Huggenberger P, Hunt B, Jeannin P-Y, Jiang G, Jones G, Jourde H, Liso IS, Liu Z, Maréchal J-C, Massei N, Mazzilli N, Mudarra M, Parise M, Pu J, Ravbar N, Sanchez LH, Santo A, Sauter M, Seidel J-L, Sivelle V, Skoglund RØ, Stevanovic Z, Wood C, Worthington S, Hartmann A (2020) Global karst springs hydrograph dataset for research and management of the world’s fastest-flowing groundwater. Sci Data 7:59 Karmann, I, Koit, O, Kordilla, J, Labat, D, Ladouche, B. https://doi.org/10.1038/s41597-019-0346-5 Zhang Q, Xu C-Y, Chen YD, Ren L (2011) Comparison of evapotranspiration variations between the Yellow River and pearl river basin, china. Stoch Environ Res Risk Assess 25:139–150. https://doi.org/10.1007/s00477-010-0428-6 Zhao Y, Deng Q, Lin Q, Zeng C, Zhong C, Sullivan PF, Stokes MC, McMillan CK, Weintraub MN (2020) Cadmium source identification in soils and high-risk regions predicted by geographical detector method. Environ Pollut 263:114338. https://doi.org/10.1016/j.envpol.2020.114338 Zuo R, Zheng S, Liu X, Wu G, Wang S, Wang J, Liu J, Huang C, Zhai Y (2023) Groundwater table fluctuation: A driving force affecting nitrogen transformation in nitrate-contaminated groundwater. J Hydrol 621:129606. https://doi.org/10.1016/j.jhydrol.2023.129606 Wang X, Xu S, Zhan J, Ma JW, Liu Junxin L, Lin G Xuesong (2021) Analysis of the effect of two-stage aerobic co-composting of rural organic solid waste. J Environ Eng 15:1708–1715 Hu QQ (2024) Pollution characteristics and risk assessment of persistent organic pollutants in typical karst industrial areas (D).Guilin University of Technology, China. https://doi.org/10.27050/d.cnki.gglgc.2024.000488 Xue JY, Liu WT, Jiang XY, Zhao ZH, Zhang L, Cai YG, Wang XL (2019) Ammonia nitrogen cycling rate and influencing factors of water bodies in the middle reaches of the Yangtze River coastal zone. Journal of Yangtze River Basin Resources and Environment 28, 2735–2742. https://doi.org/CNKI:SUN:CJLY.0.2019-11-020 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 16 May, 2026 Reviews received at journal 24 Apr, 2026 Reviewers agreed at journal 24 Apr, 2026 Reviewers agreed at journal 24 Apr, 2026 Reviewers agreed at journal 24 Apr, 2026 Reviewers agreed at journal 23 Apr, 2026 Reviewers invited by journal 10 Feb, 2026 Editor assigned by journal 27 Jan, 2026 Submission checks completed at journal 27 Jan, 2026 First submitted to journal 23 Jan, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8680219","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":590330489,"identity":"3196b481-6360-4318-9df8-af019d93d8dd","order_by":0,"name":"Zupeng Wan","email":"","orcid":"","institution":"Guilin University of Technology","correspondingAuthor":false,"prefix":"","firstName":"Zupeng","middleName":"","lastName":"Wan","suffix":""},{"id":590330490,"identity":"7f5734c2-da94-4d53-ae75-ccaa0b8ab689","order_by":1,"name":"Baojun Liu","email":"","orcid":"","institution":"Hubei College of Water Resources and Hydropower Technology","correspondingAuthor":false,"prefix":"","firstName":"Baojun","middleName":"","lastName":"Liu","suffix":""},{"id":590330491,"identity":"716649d3-a29f-47ce-a397-8c3ca7f6457a","order_by":2,"name":"Wenwen Chen","email":"","orcid":"","institution":"Guilin University of Technology","correspondingAuthor":false,"prefix":"","firstName":"Wenwen","middleName":"","lastName":"Chen","suffix":""},{"id":590330492,"identity":"e730ba6b-9a21-4143-9ee6-cb1e697b6ac7","order_by":3,"name":"Yingjie Chen","email":"","orcid":"","institution":"South China University of Technology","correspondingAuthor":false,"prefix":"","firstName":"Yingjie","middleName":"","lastName":"Chen","suffix":""},{"id":590330493,"identity":"9d0a2596-91bb-4280-b113-64b9dbfd644e","order_by":4,"name":"Xiaoyu Yan","email":"","orcid":"","institution":"Changhu Ecological Administration Bureau","correspondingAuthor":false,"prefix":"","firstName":"Xiaoyu","middleName":"","lastName":"Yan","suffix":""},{"id":590330494,"identity":"71ee2a7b-2476-49cd-94e7-b73de484e3a9","order_by":5,"name":"Honghu Zeng","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8ElEQVRIiWNgGAWjYJACZhBhwMB8gOEBmJ9AtBa2BKBiA5K08BgQp0W+vffw64KKO3bbJXI+v0hs+8PAz55jwPBzB24tjD3n0qxnnHmWvHNG7jaLxDYDBsmeNwaMvWfwOEoix8yYt+1wssGN3G0GIC0GN3IMmBnbcGthk38D05LzDKzFnpAWHgke48dALXZALcwPwLZIENAiwZNjxsxz5nCCwZlnZgwJ54x5JM48KzjYi0eLfPsZ4888FYftDY4nP/7woUxOjr89eeODn3i0gLwjASQSG6AMHpDQAbwagIH2AUjYwxijYBSMglEwCjAAAI5dUVAUtChMAAAAAElFTkSuQmCC","orcid":"","institution":"Guilin University of Technology","correspondingAuthor":true,"prefix":"","firstName":"Honghu","middleName":"","lastName":"Zeng","suffix":""}],"badges":[],"createdAt":"2026-01-23 14:38:54","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8680219/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8680219/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":102586020,"identity":"e2432f84-5c7a-4b8f-89f7-c5c8495996f4","added_by":"auto","created_at":"2026-02-13 10:12:32","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":836624,"visible":true,"origin":"","legend":"\u003cp\u003eHydrogeological profile of the Lipu River Basin\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-8680219/v1/24b6a5cdc80e5be76d393649.png"},{"id":102586047,"identity":"587f216a-356c-4555-a201-fa54269c9676","added_by":"auto","created_at":"2026-02-13 10:12:51","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":332528,"visible":true,"origin":"","legend":"\u003cp\u003eConcentrations of (a) nitrogen species and (b) total nitrogen (TN) in river water and groundwater during summer and winter seasons. SR, summer river water; WR, winter river water; SG, summer groundwater; WG, winter groundwater. Significance levels: *p \u0026lt; 0.05; **p \u0026lt; 0.01.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-8680219/v1/eb4e7dea4fa286729c6565bf.png"},{"id":102586018,"identity":"7015b33d-63a2-4c38-b289-8dec7437c03a","added_by":"auto","created_at":"2026-02-13 10:12:31","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":273878,"visible":true,"origin":"","legend":"\u003cp\u003eSpatial and temporal variations of nitrogens in river water: (a) NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e−\u003c/sup\u003e-N, (b) NH\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e+\u003c/sup\u003e-N, and (c) NO\u003csub\u003e2\u003c/sub\u003e⁻-N concentrations classified by pollutant discharge level and land use type; (d) proportions of nitrogen species relative to TN. SW, sanitary wastewater, IW; industrial wastewater; OT, other types.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-8680219/v1/e047b41ef2959a909a15b87d.png"},{"id":102586034,"identity":"3a8fba25-03f7-4500-8148-6906fa5c41cf","added_by":"auto","created_at":"2026-02-13 10:12:46","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":272700,"visible":true,"origin":"","legend":"\u003cp\u003eSpatial and temporal variations of nitrogens in groundwater: (a) NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e−\u003c/sup\u003e-N, (b) NH\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e+\u003c/sup\u003e-N, and (c) NO\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e-\u003c/sup\u003e-N concentrations classified by pollutant discharge level and land use type; (d) proportions of nitrogen species relative to TN.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-8680219/v1/bcb6cb8a5ac083f07f7a9d61.png"},{"id":102586016,"identity":"97c01d52-993e-411d-936c-af53020b4b4c","added_by":"auto","created_at":"2026-02-13 10:12:31","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1255056,"visible":true,"origin":"","legend":"\u003cp\u003eSpatial distribution of nitrogen in river water/groundwater during summer/winter season. (a) is the river water in summer, (b) is the river water in winter, (c) is the groundwater in summer, and (d) is the groundwater in winter\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-8680219/v1/4da0921676e8ccb2fdd3b302.png"},{"id":102586044,"identity":"55cff999-3355-4c64-9702-53aa325fea68","added_by":"auto","created_at":"2026-02-13 10:12:50","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":207590,"visible":true,"origin":"","legend":"\u003cp\u003eRelationships between logarithmic conversion concentrations of NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e−\u003c/sup\u003e and δ\u003csup\u003e15\u003c/sup\u003eN and δ\u003csup\u003e18\u003c/sup\u003eO in the Lipu River Basin. Shaded area represents the confidence interval of the expected value (a)(b) is the relationship between the logarithmic conversion concentration of NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e−\u003c/sup\u003e and δ\u003csup\u003e15\u003c/sup\u003eN, (c)(d) is the relationship with δ\u003csup\u003e18\u003c/sup\u003eO\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-8680219/v1/18413b0aa37d12546bbb3ec6.png"},{"id":102586035,"identity":"659b8550-92a7-4e09-a505-d1f8ec993daa","added_by":"auto","created_at":"2026-02-13 10:12:46","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":128208,"visible":true,"origin":"","legend":"\u003cp\u003e(a) Relationship between nitrate (NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e−\u003c/sup\u003e) and electrical conductivity. (b) Relationship between NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e−\u003c/sup\u003e and Cl\u003csup\u003e−\u003c/sup\u003e.\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-8680219/v1/db0671020c3393a70ea26ab0.png"},{"id":102586038,"identity":"984cd794-93a6-4d70-914b-16f34392c432","added_by":"auto","created_at":"2026-02-13 10:12:48","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":148737,"visible":true,"origin":"","legend":"\u003cp\u003eDistribution characteristics of δ\u003csup\u003e15\u003c/sup\u003eN and δ\u003csup\u003e18\u003c/sup\u003eO values in the water bodies of the study area. (a) represents the distribution of river and groundwater, (b) represents the distribution of different pollution types, (c) represents the proportion of pollution sources).\u003c/p\u003e","description":"","filename":"8.png","url":"https://assets-eu.researchsquare.com/files/rs-8680219/v1/169128ca974f77e7dd63bdee.png"},{"id":102747286,"identity":"3de34de5-9982-4dee-8ca9-1d8324618ae0","added_by":"auto","created_at":"2026-02-16 09:04:24","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4457799,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8680219/v1/5c19dfd1-a042-4e79-9700-af8076c12d61.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Probabilistic source apportionment and quantification of nitrate contamination in a karst aquifer system: Revealed by stable isotopic and hydrochemical proxies","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003ePollution control strategies established for porous-media dominated basins are often inadequate for highly heterogeneous karst environments (Guo et al. \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). When nitrogen pollution from industrial and agricultural development overlays the unique surface-subsurface hydrological structure of karst basins, pollutant transport deviates from traditional model predictions (P. Zhang et al. \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). A highly permeable karst conduit network enables refractory organic matter in industrial wastewater and agricultural nitrogen to bypass the slow degradation process in the soil zone for rapid and deep transboundary transport (G\u0026uuml;ven et al. \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Instead of attenuating pollution, this geological feature may intensify its transformation and induce biotoxicity through complex physicochemical processes, with accelerated and amplified ecological effects (Wells et al. 2018).\u003c/p\u003e \u003cp\u003eExisting studies have focused on nitrogen source apportioning in non-karst areas (Pandit et al. 2022). However, the rapid surface-subsurface exchange in karst systems limits the applicability of traditional watershed models (e.g, SWAT, HEC-HMS), which struggle to accurately characterize heterogeneous flow paths and nitrogen transport processes, thereby undermining the reliability of source apportionment analyses (Iriawan H et al.2021). Consequently, accurate quantification requires a dynamic traceability network integrating isotope techniques and hydrogeological technologies of nitrogen transport pathways in karst systems, identification of key pollution sources, and analysis of their spatiotemporal differentiation patterns. To address these challenges, this study targets the Lipu River basin, where high nitrate concentrations have been documented yet source contributions remain unquantified, employing dual nitrate isotopes coupled with hydrochemical parameters to trace nitrogen sources and transformation pathways in this representative karst system. Isotope and receptor models quantitatively determined industrial and domestic pollution source contributions, advancing from static source tracing to dynamic process analysis.\u003c/p\u003e \u003cp\u003eAquatic nitrogen exists primarily as NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e, NO\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e and NH\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e+\u003c/sup\u003e. High mobility and proportions of NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e in karst areas make it a key indicator of pollution transport (Zuo et al. \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). In addition, high NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e levels are associated with methemoglobinemia in humans and aquatic ecosystem degradation (Yan et al. \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e pollution sources in karst basins include point and river sources. Point sources include mainly industrial parks and landfills. In addition, NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e concentrations and sources are affected by various nitrogen transformations. For example, higher NH\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e+\u003c/sup\u003e concentrations can promote nitrification (Ma et al. \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) and inhibit NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e removal by assimilation, leading to NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e accumulation in karst basins. Excessive nitrogen fertilizer application in agricultural areas is typically considered the main source of NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e in the basin (Wang et al. \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), whereas point-source pollution is considered the main source of NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e pollution in urbanized areas (Lai et al. \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). However, studies have demonstrated that manure, sewage, and soil nitrogen are the major contributors to groundwater NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e pollution in agricultural areas (Cui et al. \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), suggesting that rapid urbanization and agricultural activities are altering nitrogen pollution sources and contributions in river basins. Integrating multi-source data with models facilitates precise tracing and quantitative assessment of the dynamic load of each pollution source under the dual drive of nature\u0026ndash;human enabling the transition from fuzzy attribution to precise localization.\u003c/p\u003e \u003cp\u003eNO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e sources largely reflect basin nitrogen sources (Iqbal et al. \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Schaefer et al. \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). Linking land use types to cations and anions has been a traditional approach to identifying NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e sources and nitrogen cycling processes (Biddau et al. \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), The NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e stable isotope method (δ\u003csup\u003e15\u003c/sup\u003eN-NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e and δ\u003csup\u003e18\u003c/sup\u003eO-NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e) provides meaningful insights into NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e sources and nitrogen conversion processes because different NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e sources exhibit unique isotopic characteristics (Xue et al. \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). The NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e stable isotope method requires minimal auxiliary information, offers high accuracy and operational simplicity, and directly identifies of NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e sources (Yang et al. \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Generally, δ\u003csup\u003e15\u003c/sup\u003eN-NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e is the highest in M\u0026amp;S (manure and sewage) and δ\u003csup\u003e18\u003c/sup\u003eO-NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e is the highest in AD (Atmospheric deposition), whereas \u003csup\u003e15\u003c/sup\u003eN-NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e is the lowest in mineral fertilizers, followed by in SN (soil nitrogen), which facilitates tracing of NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e sources in aquatic environments (Wu P et al. \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). However, overlappingNO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e isotope signatures from different sources limit the accuracy of the dual-isotope approach for source apportionment, such as nitrogen fertilizer to soil nitrogen, soil nitrogen to manure and sewage, mixing of multiple NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e sources, and NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e isotope fractionation during biogeochemical cycles (Lin et al. \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Consequently, numerous studies have sought to identify NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e sources and transformation processes by integrating NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e isotopes with complementary tracers (e.g, δ\u003csup\u003e34\u003c/sup\u003eS-SO\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e2\u0026minus;\u003c/sup\u003e, δ\u003csup\u003e11\u003c/sup\u003eB), hydrochemical indicators (e.g, Cl\u003csup\u003e\u0026minus;\u003c/sup\u003e, Na\u003csup\u003e+\u003c/sup\u003e), environmental parameters (e.g, dissolved oxygen), and multivariate statistical methods (e.g, redundancy analysis) (Grechanik et al. \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Li et al. \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThis study integrated a Bayesian isotope mixing model with NO\u003csub\u003e3\u003c/sub\u003e⁻ stable isotope analysis. Additional covariates (hydrochemical parameters and land use types) enhanced source apportionment accuracy compared with traditional mixing models. The hydrological status of the southwestern karst basin is complex, with intense agricultural and human activities. Groundwater, rainfall, and river water are key sources of river recharge. Towns and farmlands are scattered around karst basin (Chang et al. \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Therefore, karst basins are typical areas of mixed land use (Yi et al. \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). Frequent seasonal groundwater-river water exchange results in severe groundwater nitrogen pollution, with average nitrogen and NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e concentrations reaching 24.35 and 15.15 mg/L (Boumaiza et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), respectively, posing a major threat to lake water quality and safety. However, few studies have focused on nitrogen sources, transformation, and contributions of nitrogen sources in karst basins, potentially masking dominant nitrogen sources and pollution characteristics in rivers and groundwater.\u003c/p\u003e \u003cp\u003eTo address these research gaps, we investigated inorganic nitrogen pollution in river water and groundwater in the Lipu River Basin within Lipu County, considering domestic sewage, industrial sewage, and other pollution sources. By integrating NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e stable isotope (δ\u003csup\u003e15\u003c/sup\u003eN and δ\u003csup\u003e18\u003c/sup\u003eO) analysis with water chemistry data, meteorological parameters, environmental factors, and the Bayesian isotope mixing (SIAR) model, the objectives of this study are to: (1) clarify the distribution characteristics of DIN species under the influence of various discharge types in a typical karst area; (2) investigate the spatiotemporal distribution of nitrogen-containing pollutants in river water and groundwater and their correlations with comprehensive hydrochemical indicators; and (3) identify the sources of NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e based on nitrogen distribution characteristics and the SIAR model. These findings provide a scientific basis for evaluating nitrogen dynamics and developing management strategies for nitrogen pollution control under various discharge scenarios in karst regions.\u003c/p\u003e"},{"header":"1 Materials and methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Description of the study area\u003c/h2\u003e \u003cp\u003eThe study area is located in the northeastern part of Guangxi Zhuang Autonomous Region, in southern part of Guilin City, with Lipu City as the main administrative division. It includes the western part of Pingle County, extending to the west banks of the Lijiang and Guijiang Rivers (110\u0026deg;12\u0026prime;6\u0026Prime;\u0026ndash;110\u0026deg;40\u0026prime;57\u0026Prime; east longitude, 24\u0026deg;20\u0026prime;54\u0026Prime;\u0026ndash;24\u0026deg;39\u0026prime;27\u0026Prime; north latitude). The climate of the study area is characterized by a long, hot, and rainy summer, a warm and occasionally cold winter, distinct summer and winter, an average monthly temperature of 18.9\u0026deg;C over many years, and an average annual rainfall of 1441.5 mm, mainly concentrated in April\u0026ndash;August. The submerged aquifer in the area is 0.5\u0026ndash;3.5 meters deep. The lithology of the area is composed mainly of clastic, carbonate, igneous, and loose rocks. The Quaternary Holocene Q\u003csub\u003e4\u003c/sub\u003e is located on the upper terrace, with pore aquifers and partial pressure. The northern part of the basin overlies the Lower Carboniferous stages of the Datangdian and Yanguandian C\u003csub\u003e1\u003c/sub\u003e, which contain fissure cave water with a small flow of springs (Zhang et al. \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cdiv id=\"Sec4\" class=\"Section3\"\u003e \u003ch2\u003e2.1.1 Watershed monitoring\u003c/h2\u003e \u003cp\u003eConcentrated sampling was conducted during the summer and winter seasons in 2022, as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, Fourteen sampling points were selected along the mainstream of the Lipu River (uniformly distributed from the upstream Tianjingling to the downstream Guijiang confluence point), including 13 sampling points from the six major tributaries of the Pulu, Dumo, Datang, Xinping, Limu, and Maling Rivers. Twelve groundwater sampling points were selected based on the direction of groundwater movement and underlying hydrogeological conditions.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section3\"\u003e \u003ch2\u003e2.1.2 Sample collection\u003c/h2\u003e \u003cp\u003eBased on the hydrogeological conditions of the study area and field research, points near various sewage outlets on the main and tributary streams of the Lipu Basin were selected to cover different hydrodynamic zones (recharge, runoff, and discharge zones) in the karst basin and river and groundwater samples were collected. Before sampling, the brown glass bottles were rinsed three times with ultrapure water, rinsed three times with sample water, sealed with aluminum foil, and fastened with caps. Parallel samples were set at a ratio of no less than 10% of the total monitoring points for each batch of sampling to ensure accuracy and reliability of the monitoring data.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"2 Sample preparation and analysis","content":"\u003cdiv id=\"Sec7\" class=\"Section3\"\u003e \u003cdiv class=\"Heading\"\u003e2.2.1 Detection of inorganic carbon and nitrogen compounds\u003c/div\u003e \u003cp\u003eImmediately after the samples were transported to the laboratory, NH\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e+\u003c/sup\u003e-N, NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N, NO\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N, and inorganic carbon (DIC, including CO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e2\u0026minus;\u003c/sup\u003e and HCO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e) were measured. The remaining samples were filtered using a 0.45-\u0026micro;m filter membrane (Tianjin Jinteng, China) and stored in the dark at 0\u0026ndash;4\u0026deg;C. Among them, the DIN contents were detected using the JH-TD401 series multi-parameter water quality analyzer based on spectrophotometry principles. DIC was detected using total alkalinity test kits (10\u0026ndash;200 mg/L, Lu Heng Bio, China).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section3\"\u003e \u003cdiv class=\"Heading\"\u003e2.2.2 Hydrochemical detection and data analysis\u003c/div\u003e \u003cp\u003eThe anion and cation indicators to be tested included K\u003csup\u003e+\u003c/sup\u003e, Na\u003csup\u003e+\u003c/sup\u003e, Ca\u003csup\u003e2+\u003c/sup\u003e, Mg\u003csup\u003e2+\u003c/sup\u003e, SO\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e2\u0026minus;\u003c/sup\u003e, and Cl\u003csup\u003e\u0026minus;\u003c/sup\u003e. The samples to be tested were filtered using a 0.22-\u0026micro;m filter membrane and stored away from light at 4\u0026deg;C. Cationic concentrations were determined using an inductively coupled plasma emission spectrometer (Optima 8000, USA) in accordance with the method specified in the National Environmental Protection Standard (HJ 776\u0026ndash;2015, China). Anions were detected using ion chromatography (DIONEX ICS-2100, USA). Anions was detected using the method specified in the National Environmental Protection Standard (HJ 84-2016, China). Since the indicator data group consists of multiple measurements of the same object under different conditions and the data are not normal, the Friedman test is used to analyze the significance of the data.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Isotope sample collection and analysis\u003c/h2\u003e \u003cdiv id=\"Sec10\" class=\"Section3\"\u003e \u003ch2\u003e2.3.1 Isotope sample collection\u003c/h2\u003e \u003cp\u003eRiver water samples were collected onsite using PET sample bottles. First, they were filtered through a 0.45 \u0026micro;m filter membrane to remove various suspended solids and microorganisms. Subsequently, the water samples were stored at temperatures below 4\u0026deg;C. The denitrifying bacteria method was used to determine nitrogen and oxygen isotopes in the samples. The samples to be tested were filtered through a 0.22 \u0026micro;m cartridge filter membrane, and the nitrate concentration was measured before freezing and storage. An appropriate amount of the denitrifying bacterial strain was selected and plated. Afterward, it was incubated at 30\u0026deg;C for 3 days. The monoclonal strain obtained from the culture was obtained and incubated in a test tube containing 6 mL of TSB medium for 8 h in a constant temperature shaker at 32\u0026deg;C and 180 r/min. The remaining strain was stored at 4\u0026deg;C.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section3\"\u003e \u003ch2\u003e2.3.2 Pre-processing and on-machine operation\u003c/h2\u003e \u003cp\u003eCulture medium (0.2 mL) was drawn and added to 10 mL Tryptic Soy Broth (TSB) culture medium and incubated under the same conditions for 16 h. Finally, 10 mL of the culture medium was transferred to 238 mL of TSB culture medium and incubated under the same conditions for 48 h. After bacterial culture expansion was complete, a colorimetric test was performed. On the next day, 0.2 mL of 10 mol/L NaOH was added with a syringe to terminate the reaction and absorb CO\u003csub\u003e2\u003c/sub\u003e, and then the NO\u003csub\u003e2\u003c/sub\u003e gas generated by the reaction was analyzed on the machine.\u003c/p\u003e \u003cp\u003eHydrogen and oxygen isotopes were measured using a liquid water isotope laser spectrometer. The \u003csup\u003e13\u003c/sup\u003eC stable isotope was commissioned to the Karst Geological Resources and Environment Supervision and Testing Center of the Ministry of Land and Resources (Institute of Karst Geology, Chinese Academy of Geological Sciences, Guilin) for analysis and testing.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Nitrogen magnification slope and SIAR model\u003c/h2\u003e \u003cp\u003eNitrogen Magnification Slope (NMS) is estimated by linear regression of pollutant concentrations and δ\u003csup\u003e15\u003c/sup\u003e N with logarithmic transformation, as shown in Eq.\u0026nbsp;1 (Lavoie et al. \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2013\u003c/span\u003e):\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(\\:\\text{log}{\\text{C}}_{{\\text{NO}}_{{\\text{3}}^{\\text{-}}}}\\text{= }\\)\u003c/span\u003e \u003c/span\u003eδ\u003csup\u003e15\u003c/sup\u003eN(b)\u0026thinsp;+\u0026thinsp;a (1)\u003c/p\u003e \u003cp\u003ewhere the slope (b) represents the NMS. A positive slope (TMS\u0026thinsp;\u0026gt;\u0026thinsp;0) indicates increased N accumulation. A Bayesian stable isotope mixing model was used to quantify the contributions of different nitrate sources to the Lipu River Basin water system. The model was implemented using the SIAR (stable isotope analysis in R) package and can be represented as follows (Ji X et al, 2022):\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"No\" id=\"Taba\" border=\"1\"\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{X}}_{\\text{ij}}\\text{=}\\sum\\:{\\text{p}}_{\\text{k}}\\left({\\text{S}}_{\\text{ij}}\\text{+}{\\text{C}}_{\\text{ij}}\\right)\\text{+}{\\text{\u0026epsilon;}}_{\\text{ij}}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(2)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{S}}_{\\text{ij}}\\text{=}\\text{N}\\left({\\text{\u0026mu;}}_{\\text{ij}}\\text{,}{\\text{\u0026omega;}}_{\\text{jk}}^{\\text{2}}\\right)\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(3)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{C}}_{\\text{ij}}\\text{=\\:}\\text{N}\\left({{\\lambda}^{{\\prime\\:}}}_{\\text{ij}}\\text{,}{\\text{\u0026omega;}{\\text{\u0026tau;}}^{{\\prime\\:}}}_{\\text{jk}}^{\\text{2}}\\right)\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(4)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\text{\u0026epsilon;}}_{\\text{ij}}\\text{=}\\text{N}\\left(\\text{0,}{\\text{\u0026sigma;}}_{\\text{j}}^{\\text{2}}\\right)\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(5)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eIn the above formula (Eq.\u0026nbsp;1\u0026ndash;5), X\u003csub\u003eij\u003c/sub\u003e is the JTH isotope value of the i-th mixture, i\u0026thinsp;=\u0026thinsp;1,2,3,... N; j\u0026thinsp;=\u0026thinsp;1,2,3,... J; P\u003csub\u003ek\u003c/sub\u003e is the proportion of the KTH source that needs to be estimated using the SIAR model; S\u003csub\u003eij\u003c/sub\u003e is the value of the JTH isotope from the i-th source, following a normal distribution with a mean of \u0026micro; and a variance of ω; C\u003csub\u003eij\u003c/sub\u003e is the fractionation coefficient of the JTH isotope from the i-th source, following a normal distribution with a mean of λ 'and a variance of τ'; ε\u003csub\u003eij\u003c/sub\u003e is the residual error, representing the variance that cannot be quantified among the other mixtures, with the mean and variance typically being 0.\u003c/p\u003e \u003cp\u003eTo estimate the proportional contribution of NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N sources, the present study considered two isotopes (j\u0026thinsp;=\u0026thinsp;2) (\u003cem\u003eδ\u003c/em\u003e\u003csup\u003e15\u003c/sup\u003eN, \u003cem\u003eδ\u003c/em\u003e\u003csup\u003e18\u003c/sup\u003eO) and four major potential NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N sources (nitrogen fertilizer, soil organic matter, manure, and sewage). Four potential pollution sources were identified: fertilizer, soil, manure, and sewage. The nitrogen and oxygen isotope compositions of the different pollution sources were determined based on previous studies, and the model will be refined further when superior data are obtained. The SIAR model established in the present study was based on R v4.1.3 (R Foundation for Statistical Computing, Vienna, Austria), operated with RStudio 2022.02.0-443, and the code and script were written with reference to the Bayesian mixed model MixSIAR written by Stock.\u003c/p\u003e \u003c/div\u003e"},{"header":"3 Results and discussion","content":"\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Basic nitrogen contents\u003c/h2\u003e \u003cp\u003eThe detection rate of DIN for river water was 100%, with TN concentrations ranging from 0.24 to 16.11 mg/L (average 5.02\u0026thinsp;\u0026plusmn;\u0026thinsp;4.04 mg/L), which were (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.05) lower than that in groundwater (6.11\u0026thinsp;\u0026plusmn;\u0026thinsp;5.01 mg/L). Among them, the average content of the DIN in the river water from high to low was NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N (4.27\u0026thinsp;\u0026plusmn;\u0026thinsp;3.55 mg/L), NH\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e+\u003c/sup\u003e-N (0.68\u0026thinsp;\u0026plusmn;\u0026thinsp;0.42 mg/L), NO\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N (0.06\u0026thinsp;\u0026plusmn;\u0026thinsp;0.04 mg/L), the DIN contents in groundwater, from high to low, were NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N (5.75\u0026thinsp;\u0026plusmn;\u0026thinsp;4.82 mg/L), NH\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e+\u003c/sup\u003e-N (0.36\u0026thinsp;\u0026plusmn;\u0026thinsp;0.28 mg/L), NO\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N (0.07\u0026thinsp;\u0026plusmn;\u0026thinsp;0.05 mg/L). Approximately 15% of the water samples had NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N concentrations exceeding the natural threshold for drinking water (10 mg/L; WHO and China). The NH\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e+\u003c/sup\u003e-N concentration level is significantly (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05) higher than that in large rivers with good water quality, such as the middle reaches of the Yangtze River (0.24 mg/L) and the Pearl River (0.33 mg/L) (Zhang et al. \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Xue et al. \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), and the water quality is roughly the same as that of the main stream of the Yellow River, China, which is greatly affected by local industrial and agricultural activities. Compared with karst river basins abroad, the Lipu River Basin is much better (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05) than the Silver Spring in Florida in USA (1.55 mg/L) and the Coesle River Basin in France (2.34 mg/L). In summary, the water quality demonstrates a higher contamination risk compared to major rivers both in China and internationally, reflecting clear evidence of human impact.\u003c/p\u003e \u003cp\u003eFigure\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e presents the concentrations of nitrogen in river and groundwater samples from the Lipu River Basin. In Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea, the nitrate nitrogen in the river water during summer (average 2.29\u0026thinsp;\u0026plusmn;\u0026thinsp;1.02 mg/L) was significantly (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05) lower than that of the other DIN, while there was no significant difference in the four types of ammonia nitrogen and nitrite nitrogen (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.05), indicating that nitrate nitrogen dominates the TN and its seasonal variation is relatively large. 2b showed that the TN of river water in summer (2.95\u0026thinsp;\u0026plusmn;\u0026thinsp;1.48 mg/L) was significantly (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05) lower than that of river water in winter (7.09\u0026thinsp;\u0026plusmn;\u0026thinsp;6.59 mg/L), while there was no significant difference in the TN samples of groundwater in different seasons. This indicates that karst aquifer systems buffer and homogenize TN, while river water responds more directly and sensitively to seasonal human activities and natural processes. This phenomenon may be related to well-developed faults and folds in the western section of the Nanling Tectonic Belt, where the Lipu Basin is located (Hu, \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), providing a preferred channel for the rapid migration of karst groundwater.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cdiv id=\"Sec15\" class=\"Section3\"\u003e \u003ch2\u003e3.1.1 Nitrogen content in river water\u003c/h2\u003e \u003cp\u003eDuring the summer, the concentration of NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N in the river water ranging from 0.24 to 4.16 (average 2.29\u0026thinsp;\u0026plusmn;\u0026thinsp;2.01 mg/L), NH\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e+\u003c/sup\u003e-N (0.64\u0026thinsp;\u0026plusmn;\u0026thinsp;0.43 mg/L), NO\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N (0.03\u0026thinsp;\u0026plusmn;\u0026thinsp;0.01 mg/L), with a relative content of 77.57, 21.68 and 0.75%. NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N concentrations ranged from 0.24 to 4.16 mg/L (6.25\u0026thinsp;\u0026plusmn;\u0026thinsp;5.65 mg/L), NH\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e+\u003c/sup\u003e-N (0.76\u0026thinsp;\u0026plusmn;\u0026thinsp;0.65 mg/L), NO\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N (0.09\u0026thinsp;\u0026plusmn;\u0026thinsp;0.05 mg/L), with a relative content of 88.03, 10.70 and 1.27%. Overall, the relative content of NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N was more than 77%, indicating that the nitrogen form entering the river was mainly NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N, and the sources were mainly nitrified or rapidly nitrified pollution sources, including fertilizer application, livestock breeding, and sewage treatment plant discharge in the study area (Yu Z et al. 2022). The average concentration of NO\u003csub\u003e3\u003c/sub\u003e⁻-N in winter (6.25 mg/L) was significantly higher (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05) than that in summer (2.29 mg/L), and the fluctuation range was larger. Meanwhile, NH\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e+\u003c/sup\u003e-N and NO\u003csub\u003e3\u003c/sub\u003e⁻-N also saw a slight increase in winter. This indicates that water bodies in winter may be affected by factors such as stronger external nitrogen input or weakened absorption by aquatic plants in summer (Friedrichs et al. \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe nitrogen concentration is relatively high near industrial land, with NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N (6.65\u0026thinsp;\u0026plusmn;\u0026thinsp;5.63 mg/L), NH\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e+\u003c/sup\u003e-N (0.75\u0026thinsp;\u0026plusmn;\u0026thinsp;0.53 mg/L), NO\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N (0.04\u0026thinsp;\u0026plusmn;\u0026thinsp;0.03 mg/L). However, there was no significant (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.05) difference in DIN concentration under industrial, agricultural and other land use types, indicating that the pollutants in the river water underwent intense mixing and homogenization. Speculative study area had a mixed connection and discharge in the river sewage collection system (Shoaib et al. \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2008\u003c/span\u003e), which led to further mixing of pollutants during transportation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section3\"\u003e \u003ch2\u003e3.1.2 Nitrogen content in groundwater\u003c/h2\u003e \u003cp\u003eThe TN concentration of groundwater in the basin (average 6.12\u0026thinsp;\u0026plusmn;\u0026thinsp;5.33 mg/L) was higher than that of river water, indicating persistent pollution seepage and accumulation. Among them, the NH\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e+\u003c/sup\u003e-N concentration in groundwater during the summer (0.34\u0026thinsp;\u0026plusmn;\u0026thinsp;0.31 mg/L) with a relative content of TN was 5.25%. NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N concentration (5.98\u0026thinsp;\u0026plusmn;\u0026thinsp;5.11 mg/L). NO\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N concentration (0.02\u0026thinsp;\u0026plusmn;\u0026thinsp;0.01 mg/L) account for 0.22% of the TN content. The concentration of NH\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e+\u003c/sup\u003e-N in groundwater during the winter (0.37\u0026thinsp;\u0026plusmn;\u0026thinsp;0.34 mg/L) with a relative content of TN was 6.44%. NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N concentration (5.36\u0026thinsp;\u0026plusmn;\u0026thinsp;5.21 mg/L) NO\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N concentrations (0.03\u0026thinsp;\u0026plusmn;\u0026thinsp;0.03 mg/L) account for 0.22% of the TN content. The concentration of NO\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N in the summer was significantly (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05) higher than in the winter, with increased rainfall in the summer and river runoff scouring dissolved NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N precursors into the vesicular zone for infiltration. In karst basins, rapid infiltration through fissure-dominated flow pathways facilitates pollutant transport to phreatic aquifers during summer, leading to elevated NO\u003csub\u003e2\u003c/sub\u003e⁻-N concentrations (Chu et al. \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N concentration (4.53\u0026thinsp;\u0026plusmn;\u0026thinsp;3.88 mg/L) with the annual average concentration in the Central High-Tech Industrial Park exceeding the natural threshold for drinking water (10 mg/L; WHO and China). IW discharge outlets include industries such as chemical, pharmaceutical, and electroplating industries, which use large amounts of raw materials such as nitrates and nitrites in production processes (Huo et al. \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Li et al. \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). The nutrients may partially remain in the wastewater, in turn increasing the concentrations of NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N in the wastewater. The average NH\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e+\u003c/sup\u003e-N in groundwater was 0.41 mg/L, which was slightly lower than the concentration in river water bodies, and the pH of water in karst basins was weakly alkaline (pH 7.5\u0026ndash;8.5). Studies on the pre-loss of NH\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e+\u003c/sup\u003e-N volatilization in rivers under alkaline conditions and the specificity of NH\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e+\u003c/sup\u003e-N volatilization in karst areas suggest that the amount of NH\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e+\u003c/sup\u003e-N entering groundwater is relatively low (Woodward et al. \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2011\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eUnlike in river water, the NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N concentration in industrial wastewater was significantly different from those in sanitary wastewater and other types, indicating that the hydrological paths of groundwater and river water were different. The main source of NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N pollution in groundwater in the basin is closely related to hanger manufacturing and specialty food processing industries that are dominant by Lipu County (M. Zhang et al. \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Nitrogen sources are converted to NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N through NH\u003csub\u003e3\u003c/sub\u003e oxidation and nitrification during wastewater treatment, resulting in an increase in discharge concentration, conversion of wastewater with high organic N from food processing, and natural oxidation of nitrite, eventually generating NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N into groundwater. This finding aligns with Wu\u0026rsquo;s probabilistic assessments (Wu et al. \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Groundwater discharge data suggest that the lack of significant differences in N\u003csub\u003e3\u003c/sub\u003e in river water does not deny the differences in the characteristics of the discharge sources themselves, but rather highlights the fundamental differences in pollution plume mixing capacity between the river river water environment and the underground aquifer environment in the region.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Spatial distribution\u003c/h2\u003e \u003cdiv id=\"Sec18\" class=\"Section3\"\u003e \u003ch2\u003e3.2.1 Nitrogen content in river water\u003c/h2\u003e \u003cp\u003eDuring the observation period, the concentration of TN in the river water (average 4.95\u0026thinsp;\u0026plusmn;\u0026thinsp;4.36 mg/L). Taking the highest proportion of nitric nitrogen as the benchmark, according to the WHO and China standards, the proportion of river monitoring points exceeding the threshold is 33.95%. The average concentrations in the tributaries of the Puling, Dumo, Datang, and Limu rivers in the basin were relatively high, and the lower reaches of the Maling River points were affected by the grain products and manufacturing processing industrial park, and the concentration of N was relatively high, increasing by 2.2 mg/L compared with that upstream of the tributaries. The concentrations of various forms of nitrogen in the main stream are NH\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e+\u003c/sup\u003e-N (0.55\u0026thinsp;\u0026plusmn;\u0026thinsp;0.38 mg/L), NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N (2.07\u0026thinsp;\u0026plusmn;\u0026thinsp;1.78 mg/L), and NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N (0.02\u0026thinsp;\u0026plusmn;\u0026thinsp;0.01 mg/L). In the tributaries, they are respectively NH\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e+\u003c/sup\u003e-N (2.21\u0026thinsp;\u0026plusmn;\u0026thinsp;1.87 mg/L), NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N (2.52\u0026thinsp;\u0026plusmn;\u0026thinsp;2.38 mg/L), and NO\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N (0.01\u0026thinsp;\u0026plusmn;\u0026thinsp;0.01 mg/L). In the main stream, the Lipu River showed a low trend in the middle reaches and a high trend in the upper and lower reaches. The main reasons are that the middle reaches fall within the urban area of Lipu County, where there are many hospitals and municipal sewage outlets, and the industrial types near DT2 are complex, covering metal processing, oil industry, rice, and flour products, etc. Strong karst development in the tributaries leads to low DO (\u0026lt;\u0026thinsp;2 mg/L), and an anoxic environment severely inhibits nitrification (Diab et al. \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e1993\u003c/span\u003e), resulting in the accumulation of NH\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e+\u003c/sup\u003e-N in the tributaries. The TN pollution levels in tributaries were higher (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05) than those in the main stream (5.11 mg/L, 2.64 mg/L), and the flow rates in tributaries were mostly over 3.5 m\u003csup\u003e3\u003c/sup\u003e/s, which is more than 30% of the flow rate of the main stream. The pollution load is relatively high, indicating that tributaries such as the Pulu and Dumo Rivers are the core sources of N pollution in the basin (Fakouri et al. \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). The river water monitoring points were mainly driven by OT and SW.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003ePoints L5, DM1, and DM2 were affected by nearby karst skylights, with higher DO (5 mg/L) and NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N concentrations of 3.05, 4.16, and 2.93 mg/L, respectively. Near PL2, L6, and DT2, there were large karst springs with NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N concentrations of 3.11, 2.97 and 3.05 mg/L respectively, all higher than the average NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N concentration (2.88 mg/L) in the basin, confirming the earlier conclusion of groundwater recharge to the river water. The monitoring data near the karst springs and skylights were significantly different from those at other points in the basin (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05). Considering the regional monitoring points were pure carbonate rock hydrogeological conditions, the retention time of pollutants around the large springs and skylights was potentially shortened (Meng et al. \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) and the natural degradation rate was relatively low.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section3\"\u003e \u003ch2\u003e3.2.2 Nitrate-nitrogen content in groundwater\u003c/h2\u003e \u003cp\u003eThe average groundwater TN concentration was 5.743 mg/ L. Based on the WHO and China standards, 18.83% of groundwater monitoring points were exceeding the threshold. Affected by industrial parks, such as metalworking, the concentration of TN at monitoring points near U10 and 11 was 2\u0026ndash;3 times higher than the overall regional concentration (Fig.\u0026nbsp;5cd)). This indicates that the IW outlet is a \"hot source area\" for triazine, which directly leads to a sudden increase in triazine concentration in the surrounding wells.\u003c/p\u003e \u003cp\u003eThe upstream agricultural area has become a significant source of groundwater NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N pollution, and the fundamental driving factor is extensive application of organic fertilizer in the Lipu Taro planting model and the subsequent nitrification process. The concentrations of NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N in the U1\u0026ndash;U3 region were 1.65\u0026ndash;3.57 mg/L higher than the average concentration of groundwater in the region. Organic fertilizers can account for 30%\u0026ndash;50% of the cultivation of Lipu taro in the region. Organic nitrogen is converted into NH\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e+\u003c/sup\u003e-N through microbial mineralization and then into NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N by nitrifying bacteria. If the application is excessive or uncomposted, the rate of nitrate formation exceeds the absorption capacity of the crop (Cui et al. \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2023\u003c/span\u003eb), resulting in accumulation of NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N. U4 and U5 were affected by nearby karst skylights, with NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N concentrations of 2.87 and 3.2 mg/L, respectively. There are large karst springs near U10, U11, and U12, with NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N concentrations of 6.1, 18.3, and 5.6 mg/L, respectively. The monitoring data near the karst spring showed no significant differences from the other points in the basin; however, the monitoring points near the karst skylight showed significant differences from those of the other points (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05). The hydrogeological types at points U10, U11, and U12 were bedrock fissure water, which was restricted by atmospheric precipitation recharge, with short underground runoff and scarce water volume (Yi et al. \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). The flow rate of the spring was generally 0.014\u0026ndash;0.325 L/s, and the average concentration of NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N at U10, U11, and U12 was relatively high (10 mg/L), which was also affected by discharge from the new materials industrial park.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Seasonal variation\u003c/h2\u003e \u003cp\u003eRiver water TN (average 2.96\u0026thinsp;\u0026plusmn;\u0026thinsp;2.36 mg/L) \u0026lt; groundwater TN (6.74\u0026thinsp;\u0026plusmn;\u0026thinsp;5.36 mg/L) during summer, and river water TN (7.12\u0026thinsp;\u0026plusmn;\u0026thinsp;6.65 mg/L) \u0026gt; groundwater TN (5.77\u0026thinsp;\u0026plusmn;\u0026thinsp;4.86 mg/L) during the winter (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e, \u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). In river water, the concentrations of NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N, NH\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e+\u003c/sup\u003e-N, and NO\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N from October to March of the following year were higher than those from July to September, mainly because of abundant precipitation during the summer and enhanced river runoff, which diluted the three N contents in the water body. During the winter, there is less precipitation and water flow slows down, and triazine accumulation in the basin leads to higher concentration. Sewage outlets within the basin also affect the distribution of nitrogen; however, the dilution and accumulation effects caused by seasonal changes are more significant. Notably, there was a significant difference in NO\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N between the wet and winters, with concentrations in the summer being approximately three times those in the winter. The unique hydrological and water quality conditions in the study area during the summer (high load input, turbidity, and anoxic microenvironment) disrupted N transformation (Zhao et al. \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) and inhibited the nitrite consumption pathway (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eb). In groundwater, the high value areas of the DIN are often concentrated at the IW monitoring point (Fig.\u0026nbsp;5cd), which reflects both the continuous influence of industrial pollution sources and the fissure-pipe type flow characteristics of karst basins. Once pollutants enter a karst aquifer through infiltration, they migrate to relatively connected channels, forming relatively stable high-value areas.\u003c/p\u003e \u003cp\u003eThe proportion of nitrogen in river water changed during the wet and winters (the proportion of NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N decreased from 88.5% to 70.2%) (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). During the summer, NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N-dominated nonpoint source pollution is prevalent, and the hydrochemical characteristics are strongly influenced by agricultural nonpoint sources, whereas the point source pollution characteristics are masked. The winter is dominated by point source pollution (with an increase of 10\u0026ndash;20 times in NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N). The water chemical characteristics reflect the direct influence of OT and IW, with a prominent contribution of point source pollution. NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N, such as in DT1, LM2, and L13, was altered from a secondary component during the summer (a) to a dominant component among the DIN sources. Groundwater did not exhibit such a trend, but the proportion of NO\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N at some monitoring points during the winter increased 2\u0026ndash;3 times compared with that in the summer. The conclusion can be explained by the denitrification process, in which the karst aquifer is in an anoxic microenvironment, which inhibits further conversion of NO\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N to NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003e3.4 Relationship between NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e and each water quality index\u003c/h2\u003e \u003cdiv id=\"Sec22\" class=\"Section3\"\u003e \u003ch2\u003e3.4.1 NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e and isotope\u003c/h2\u003e \u003cp\u003eThere was a positive correlation between the logarithmic conversion concentration of NO\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N and δ\u003csup\u003e15\u003c/sup\u003eN in SW and IW (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e), and NO\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N from manure/sewage sources in industrial and agricultural areas conformed to the characteristics of high NO\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N accompanied by high δ\u003csup\u003e15\u003c/sup\u003eN (Serna et al. \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2010\u003c/span\u003e), at which point δ\u003csup\u003e15\u003c/sup\u003eN is higher (10 to 25\u0026permil;). In contrast, an inverse relationship was observed between OT logarithmic conversion concentration and δ\u003csup\u003e15\u003c/sup\u003eN, and the dominant factor was presumed to be the very low δ\u0026sup1;⁵N value of synthetic fertilizers (such as urea, ammonium salts, nitrate fertilizers), close to the atmospheric δ\u003csup\u003e15\u003c/sup\u003eN value (0\u0026permil;), generally ranging from \u0026minus;\u0026thinsp;6\u0026permil; to +\u0026thinsp;6\u0026permil;. When fertilizers are applied in large quantities and not fully absorbed by plants or fixed in soil, the concentration of NO\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N entering the discharge outlet through runoff or leaching is very high (Sturm et al. 2011), whereas the δ\u003csup\u003e15\u003c/sup\u003eN value is relatively low.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe logarithmic conversion concentrations of NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e at SW and IW monitoring points were positively correlated with δ\u003csup\u003e18\u003c/sup\u003eO, suggesting the presence of different sources of nitrate (domestic, industrial, soil mineralization) in the water body, with significant differences in δ\u003csup\u003e18\u003c/sup\u003eO values from these sources (Brooks et al. \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2012\u003c/span\u003e), and positive correlation between NO\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N concentration and δ\u003csup\u003e18\u003c/sup\u003eO may occur after mixing. The nitrifications at SW and IW monitoring points were stronger than that at OT, the NO\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N concentration increased, and the δ\u003csup\u003e18\u003c/sup\u003eO value approached the δ\u003csup\u003e18\u003c/sup\u003eO value of the water source, so that it was positively correlated with the NO\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N concentration.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec23\" class=\"Section3\"\u003e \u003ch2\u003e3.4.2 NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e and ion\u003c/h2\u003e \u003cp\u003eThe degree of mineralization was determined based on electrical conductivity (EC) values using Cl\u003csup\u003e\u0026minus;\u003c/sup\u003e conserved tracer sources (Siddique et al. \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). The relationship between Cl\u003csup\u003e\u0026minus;\u003c/sup\u003e and NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e, and between EC and NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e, can determine whether the change in NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e concentration in water can be attributed to the mixed process of N-free transformation or denitrification. If the increase in nitrate concentration is accompanied by increases in Cl\u003csup\u003e\u0026minus;\u003c/sup\u003e and EC concentrations, the NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e concentration in the water may be affected by the mixed process of nitrogen form transformation. That is, denitrification is the main nitrogen transformation process.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe EC and Cl\u003csup\u003e\u0026minus;\u003c/sup\u003e in the water bodies of all types of sewage outlets were positively correlated (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e), and the correlation was the highest at industrial sewage outlets (R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.596), indicating that wastewater from industrial parks in the study area often contained chlorinated compounds (such as hydrochloric acid and sodium chloride). High-temperature wastewater is produced during sterilization, disinfection, cooking, and drying (Bartkowiak et al. \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). When the temperature increased, the mobility of the chloride ions increased, as did the EC. NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e and Cl\u003csup\u003e\u0026minus;\u003c/sup\u003e were positively correlated at all discharge outlets, which is a relationship that is prevalent in various pollution scenarios and reflects the pattern of synergistic changes among pollutants. The data show that the linear relationship between NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e and Cl\u003csup\u003e\u0026minus;\u003c/sup\u003e concentrations in groundwater in the basin is not significant. In addition to anthropogenic pollution sources, this may be due to karst inhomogeneity in the vertical direction, with some monitoring points under hypoxic or anoxic conditions (Lorette et al. \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), resulting in local denitrification.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec24\" class=\"Section2\"\u003e \u003ch2\u003e3.5 Source analysis of isotopes based on SIAR model\u003c/h2\u003e \u003cp\u003eThe SIAR model was used to calculate the contribution rates of the five pollution sources to NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N in the river water and groundwater in the study area. The largest source of pollution in the basin's river water was M\u0026amp;S, followed by SN, possibly because the river water was affected considerably by river runoff (Cui et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), and heavy rainfall during the summer washed the homestead soil, carrying high amounts of organic nitrogen into the river water. Studies have shown that some of the organic nitrogen is utilized by nitrification after being introduced into river water and is not transported to groundwater (Xiong et al. \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), but soil remains the second largest source of nitrogen in groundwater, which may be either transported from river water or directly released into groundwater by soil leaching. The largest sources of nitrogen in groundwater are manure and sewage. Garbage stations may use sewage to flush landfills that do not produce much runoff, most of the water will seep directly, and the process will also leach the soil, which explains the close ratio of soil to sewage sources in the industrial emissions areas (Vadas et al. \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2008\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe δ\u003csup\u003e18\u003c/sup\u003eO-NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e values of the water samples in the study area were all within the typical δ\u003csup\u003e18\u003c/sup\u003eO-NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e range produced by nitrification (Li et al. \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), indicating that nitrification is a key process for establishment of the source of NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e in river water and groundwater. In groundwater dominated by nitrification, the higher the dissolved oxygen content, the higher the δ\u003csup\u003e15\u003c/sup\u003eN-NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e value. The nitrogen and oxygen isotope fractionation produced during denitrification followed Rayleigh fractionation, and the δ\u003csup\u003e15\u003c/sup\u003eN and δ\u003csup\u003e18\u003c/sup\u003eO values of the remaining NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e satisfied the Rayleigh formula, showing that as NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e concentration decreased, both δ\u003csup\u003e15\u003c/sup\u003eN and δ\u003csup\u003e18\u003c/sup\u003eO values increased and were linearly correlated. In addition, the slope of the enrichment coefficient ratio (ε\u003csub\u003eN\u003c/sub\u003e/ε\u003csub\u003eO\u003c/sub\u003e) varied between 1.3 and 2.1. Most of the data at the monitoring points were concentrated in the left zone of ε\u003csub\u003eN\u003c/sub\u003e/ε\u003csub\u003eO\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;2.1, where the NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N concentration and the values of ni and oxygen isotopes were within a certain range, indicating that the changes in the water body were not significant and there was no obvious denitrification. The δ\u003csup\u003e15\u003c/sup\u003eN and δ\u003csup\u003e18\u003c/sup\u003eO values in groundwater were significantly inversely proportional to the NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e-N concentration, and the slope of the enrichment coefficient ratio of the two was about 1.38, suggesting denitrification in groundwater (Yuan et al. \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), combined with the fact that most of the δ\u003csup\u003e15\u003c/sup\u003eN and δ\u003csup\u003e18\u003c/sup\u003eO values in groundwater were low. The analyzed water samples exhibited suboxic conditions (DO\u0026thinsp;\u0026lt;\u0026thinsp;2 mg/L), indicating that denitrification was not intense. The δ\u003csup\u003e15\u003c/sup\u003eN and δ\u003csup\u003e18\u003c/sup\u003eO values in groundwater were slightly higher than those in river water, suggesting that groundwater was more strongly influenced by fecal matter and domestic sewage inputs. This is likely attributable to the widespread use of dry toilet systems in rural areas and the application of manure to agricultural land (Wang et al. \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe δ\u003csup\u003e18\u003c/sup\u003eO-NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e of agricultural area was higher than that of other areas, suggesting that major nitrogen biogeochemical processes have changed across different pollution types. Many studies have found that pollution type and water body characteristics largely influence the NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e isotopic characteristics (Choi et al. \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Yu et al. \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). The δ\u003csup\u003e15\u003c/sup\u003eN-NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e values observed in IA were lower than those observed in FA, mainly because SN and NF are the main sources of NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e (Kim et al. \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e1996\u003c/span\u003e). The enrichment of NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e isotopes in river water may be due to the increased consumption of nitrogen fertilizer and intensified volatilization of livestock manure during the period, as well as the evaporation of leachate from landfills, septic tanks, and domestic waste at high temperatures. However, δ\u003csup\u003e18\u003c/sup\u003eO-NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e values are influenced mainly by microbial nitrification, and not by nitrogen sources, land use, or seasonal variations. Overall, the higher δ\u003csup\u003e15\u003c/sup\u003eN-NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e in groundwater is mainly attributed to M\u0026amp;S discharge, as well as higher population density and garbage leachate.\u003c/p\u003e \u003c/div\u003e"},{"header":"4 Conclusion","content":"\u003cp\u003eThis study establishes a novel probabilistic framework for nitrate source apportionment in karst aquifer systems by integrating dual stable isotope analysis (δ\u003csup\u003e15\u003c/sup\u003eN-NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e and δ\u003csup\u003e18\u003c/sup\u003eO-NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e) with Bayesian mixing models (SIAR) and multivariate hydrochemical indicators. The key findings reveal distinct stochastic patterns in nitrogen dynamics: Groundwater TN concentrations (6.12\u0026thinsp;\u0026plusmn;\u0026thinsp;5.33 mg/L) systematically exceeded river water levels (5.02\u0026thinsp;\u0026plusmn;\u0026thinsp;4.04 mg/L), with seasonal variability amplified by karst-specific rapid recharge pathways. Spatially, tributaries adjacent to industrial zones exhibited elevated nitrogen levels, particularly near the grain processing and metal manufacturing facilities. Hydrochemical and isotopic evidence revealed source-specific signatures. Significant positive correlations between NO₃⁻ and both Cl⁻ and EC (R\u0026sup2; = 0.596) in industrial wastewater indicated high dissolved salt co-discharge with limited in-stream denitrification. The winter correlation between NO\u003csub\u003e3\u003c/sub\u003e⁻ and SO\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e2\u003c/sup\u003e⁻ confirmed co-application of sulfur- and nitrogen-containing fertilizers in agricultural areas. The δ\u003csup\u003e15\u003c/sup\u003eN-δ\u003csup\u003e18\u003c/sup\u003eO enrichment slope of 1.38, combined with δ\u003csup\u003e18\u003c/sup\u003eO-NO\u003csub\u003e3\u003c/sub\u003e⁻ values ranging from \u0026minus;\u0026thinsp;10\u0026permil; to 10\u0026permil;, demonstrates that nitrification dominates over denitrification in the basin's groundwater system\u0026mdash;a critical finding for understanding nitrogen fate in karst aquifers. SIAR modeling quantified manure and sewage as the dominant nitrate source, contributing 56.7\u0026thinsp;\u0026plusmn;\u0026thinsp;8.2% of total inputs, demonstrating the statistical robustness achievable through Bayesian inference in complex multi-source environments. Based on these findings, we recommend: (1) Implement an integrated source-pathway-receptor protection strategy strategic livestock relocation from karst-sensitive zones to consolidated feeding operations with advanced manure treatment systems (targeting 30% reduction in diffuse M\u0026amp;S sources within 5 years) and (2) given the accumulation of nitrogen load during the winter months, it is essential to collaborate with local agricultural cooperatives and food processing industries\u0026mdash;particularly those involved in taro and grain production\u0026mdash;to implement targeted nitrogen monitoring programs specifically for the winter season. This integrated isotopic-hydrochemical framework offers a transferable methodology for nitrogen source apportionment in heterogeneous karst systems.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e This work was supported by the Guangxi University Engineering Research Center for River Basin Protection and Green Development, Guilin University of Technology and the Guangxi Engineering Research Center for Comprehensive Control of Agricultural Non-point Source Pollution.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u0026nbsp;\u003c/strong\u003eThis work was funded by the National Natural Science Foundation of China (grant 52360001) and the Guangxi Basic Ability Enhancement Programme for Young and Middle-Aged Teachers (grant 2023KY0273).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u0026nbsp;\u003c/strong\u003eZupeng Wan: Conceptualization, Data curation, Form analysis, model, Writting \u0026mdash;Original draft, Writing\u0026mdash;reviewing and editing, and Funding acquisition. Baojun Liu: Data curation, Investigation. Xiaoyu Yan: Form analysis. Yingjie Chen: Conceptualization. Wenwen Chen: Resources, Methodology. Honghu Zeng\u003csup\u003e*\u003c/sup\u003e: Writing\u0026mdash;reviewing and editing, Funding acquisition.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e The authors declare no competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eBartkowiak A, Bartkowiak A, Jaworska H, Dąbkowska-Naskręt H, Rydlewska M (2020) Effect of salinity on the mobility of trace metals in soils near a soda chemical factory. J Elem 25:501\u0026ndash;512. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5601/jelem.2019.24.2.1875\u003c/span\u003e\u003cspan address=\"10.5601/jelem.2019.24.2.1875\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBiddau R, Dore E, Da Pelo S, Lorrai M, Botti P, Testa M, Cidu R (2023) Geochemistry, stable isotopes and statistic tools to estimate threshold and source of nitrate in groundwater (Sardinia, Italy). Water Res 232:119663. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.watres.2023.119663\u003c/span\u003e\u003cspan address=\"10.1016/j.watres.2023.119663\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBoumaiza L, Walter J, Chesnaux R, Huneau F, Garel \u0026Eacute;, Erostate M, Johannesson KH, Vystavna Y, Bougherira N, Bordeleau G, Stotler RL, Blarasin M, Guti\u0026eacute;rrez M, Kn\u0026ouml;ller K, Stumpp C (2022) Multi-tracer approach to understand nitrate contamination and groundwater-surface water interactions in the Mediterranean coastal area of guerbes-senhadja, algeria. J Contam Hydrol 251:104098. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jconhyd.2022.104098\u003c/span\u003e\u003cspan address=\"10.1016/j.jconhyd.2022.104098\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrooks JR, Wigington PJ, Phillips DL, Comeleo R, Coulombe R (2012) Willamette River Basin surface water isoscape (δ\u003csup\u003e18\u003c/sup\u003eO and δ\u003csup\u003e2\u003c/sup\u003eH): temporal changes of source water within the river. Ecosphere 3:1\u0026ndash;21. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1890/ES11-00338.1\u003c/span\u003e\u003cspan address=\"10.1890/ES11-00338.1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChang Y, Qi Y, Wang Z (2024) Comprehensive evaluation of IMERG, ERA5-Land and their fusion products in the hydrological simulation of three karst catchments in Southwest China. J Hydrol Reg Stud 52:101671. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ejrh.2024.101671\u003c/span\u003e\u003cspan address=\"10.1016/j.ejrh.2024.101671\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChoi J-Y, Kim S-K, Kim J-C, La G-H (2020) Utilization of nitrate stable isotopes of Chydorus sphaericus (OF M\u0026uuml;ller) to elucidate the hydrological characteristics of riverine wetlands in the Nakdong River, South Korea. J Ecol Environ 44:1. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s41610-019-0144-1\u003c/span\u003e\u003cspan address=\"10.1186/s41610-019-0144-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChu X, Ding H, Zhang X (2021) Simulation of solute transport behaviors in saturated karst aquifer system. Sci Rep 11:15614. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-021-94950-7\u003c/span\u003e\u003cspan address=\"10.1038/s41598-021-94950-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCui N, Cai M, Zhang X, Abdelhafez AA, Zhou L, Sun H, Chen G, Zou G, Zhou S (2020) Runoff loss of nitrogen and phosphorus from a rice paddy field in the east of China: Effects of long-term chemical N fertilizer and organic manure applications. Glob Ecol Conserv 22:e01011. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.gecco.2020.e01011\u003c/span\u003e\u003cspan address=\"10.1016/j.gecco.2020.e01011\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCui R, Zhang D, Hu W, Zhao X, Yan H, Liu G, Chen A (2023) Nitrogen in soil, manure and sewage has become a major challenge in controlling nitrate pollution in groundwater around plateau lakes, Southwest China. J Hydrol 620:129541. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jhydrol.2023.129541\u003c/span\u003e\u003cspan address=\"10.1016/j.jhydrol.2023.129541\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCui R, Zhang D, Hu W, Zhao X, Yan H, Liu G, Chen A (2023) Nitrogen in soil, manure and sewage has become a major challenge in controlling nitrate pollution in groundwater around plateau lakes, Southwest China. J Hydrol 620:129541. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jhydrol.2023.129541\u003c/span\u003e\u003cspan address=\"10.1016/j.jhydrol.2023.129541\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDiab S, Kochba M, Avnimelech Y (1993) Nitrification pattern in a fluctuating anaerobic-aerobic pond environment. Water Res 27:1469\u0026ndash;1475. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/0043-1354(93)90027-F\u003c/span\u003e\u003cspan address=\"10.1016/0043-1354(93)90027-F\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFakouri B, Mohamad Vali Samani J, Mohamad Vali Samani H, Mazaheri M (2022) Cost-based model for optimal waste-load allocation and pollution loading losses in river system: simulation\u0026ndash;optimization approach. Int J Environ Sci Technol 19:12103\u0026ndash;12118. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s13762-022-04422-2\u003c/span\u003e\u003cspan address=\"10.1007/s13762-022-04422-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFriedrichs MAM, St-Laurent P, Xiao Y, Hofmann E, Hyde K, Mannino A, Najjar RG, Narv\u0026aacute;ez DA, Signorini SR, Tian H, Wilkin J, Yao Y, Xue J (2019) Ocean circulation causes strong variability in the mid-atlantic bight nitrogen budget. J Geophys Res Oceans 124:113\u0026ndash;134. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2018JC014424\u003c/span\u003e\u003cspan address=\"10.1029/2018JC014424\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrechanik SV, Klymenko NA, Bunetskyi VA, Smolin SK, Zabneva OV, Nevynna LV (2024) Production of Activated Biochar from Wood Raw Materials for Water Treatment and Water Purification Applications. J Water Chem Technol 46:512\u0026ndash;523. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3103/S1063455X24050047\u003c/span\u003e\u003cspan address=\"10.3103/S1063455X24050047\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuo W, Zhang K, Liang Z, Zou R, Xu Q (2019) Electrochemical nitrogen fixation and utilization: theories, advanced catalyst materials and system design. Chem Soc Rev 48:5658\u0026ndash;5716. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1039/C9CS00159J\u003c/span\u003e\u003cspan address=\"10.1039/C9CS00159J\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eG\u0026uuml;ven D, Hanhan O, Aksoy EC, Insel G, \u0026Ccedil;okg\u0026ouml;r E (2017) Impact of paint shop decanter effluents on biological treatability of automotive industry wastewater. J Hazard Mater 330:61\u0026ndash;67. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jhazmat.2017.01.048\u003c/span\u003e\u003cspan address=\"10.1016/j.jhazmat.2017.01.048\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuo J, He G, Chen W, Hu X, Deng Q, Chen D (2019) A minireview of hydroamination catalysis: alkene and alkyne substrate selective, metal complex design. BMC Chem 13:89. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s13065-019-0606-7\u003c/span\u003e\u003cspan address=\"10.1186/s13065-019-0606-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIqbal J, Su C, Wang M, Abbas H, Baloch MYJ, Ghani J, Ullah Z, Huq ME (2022) Groundwater fluoride and nitrate contamination and associated human health risk assessment in South Punjab. Pakistan Environ Sci Pollut Res 30:61606\u0026ndash;61625. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s11356-023-25958-x\u003c/span\u003e\u003cspan address=\"10.1007/s11356-023-25958-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIriawan H, Andersen SZ, Zhang X, Comer BM, Barrio J, Chen P, Medford AJ, Stephens IEL, Chorkendorff I, Shao-Horn Y (2021) Methods for nitrogen activation by reduction and oxidation. Nat Rev Methods Primer 1:56. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s43586-021-00053-y\u003c/span\u003e\u003cspan address=\"10.1038/s43586-021-00053-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJi X, Shu L, Chen W, Chen Z, Shang X, Yang Y, Dahlgren RA, Zhang M (2022) Nitrate pollution source apportionment, uncertainty and sensitivity analysis across a rural-urban river network based on δ\u003csup\u003e15\u003c/sup\u003eN/δ\u003csup\u003e18\u003c/sup\u003eO-NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026ndash;\u003c/sup\u003e isotopes and SIAR modeling. J Hazard Mater 438:129480. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jhazmat.2022.129480\u003c/span\u003e\u003cspan address=\"10.1016/j.jhazmat.2022.129480\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKim CS, Sandretto C, Hostetler J (1996) Effects of Farmer Response to Nitrogen Fertilizer Management Practices on Groundwater Quality. Water Resour Res 32:1411\u0026ndash;1415. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/96WR00491\u003c/span\u003e\u003cspan address=\"10.1029/96WR00491\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLai Y, Yang H, Qiu F, Dang Z, Luo Y (2023) Can Rural Industrial Integration Alleviate Agricultural Non-Point Source Pollution? Evidence from Rural China. Agriculture 13:1389. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/agriculture13071389\u003c/span\u003e\u003cspan address=\"10.3390/agriculture13071389\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLavoie RA, Jardine TD, Chumchal MM, Kidd KA, Campbell LM (2013) Biomagnification of Mercury in Aquatic Food Webs: A Worldwide Meta-Analysis. Environ Sci Technol 47:13385\u0026ndash;13394. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1021/es403103t\u003c/span\u003e\u003cspan address=\"10.1021/es403103t\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi H, Tian R, Zhang Y (2025) Electrochemically superior ammonia production from electroplating wastewater over oxygen-incorporated antiperovskite. J Colloid Interface Sci 683:49\u0026ndash;57. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jcis.2024.12.122\u003c/span\u003e\u003cspan address=\"10.1016/j.jcis.2024.12.122\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi X, Tang C, Cao Y, Li X (2019) Carbon, nitrogen and sulfur isotopic features and the associated geochemical processes in a coastal aquifer system of the Pearl River Delta, China. J Hydrol 575:986\u0026ndash;998. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jhydrol.2019.05.092\u003c/span\u003e\u003cspan address=\"10.1016/j.jhydrol.2019.05.092\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi Y, Yang J, Wen J (2023) Entropy-based redundancy analysis and information screening. Digit Commun Netw 9:1061\u0026ndash;1069. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.dcan.2021.12.001\u003c/span\u003e\u003cspan address=\"10.1016/j.dcan.2021.12.001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLin H, Yan S, Song B, Wang Z, Sun L (2019) Discrimination of aged rice using colorimetric sensor array combined with volatile organic compounds. J Food Process Eng 42:e13037. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/jfpe.13037\u003c/span\u003e\u003cspan address=\"10.1111/jfpe.13037\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLorette G, Sebilo M, Buquet D, Lastennet R, Denis A, Peyraube N, Charriere V, Studer J-C (2022) Tracing sources and fate of nitrate in multilayered karstic hydrogeological catchments using natural stable isotopic composition (δ\u003csup\u003e15\u003c/sup\u003eN-NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026ndash;\u003c/sup\u003e and δ\u003csup\u003e18\u003c/sup\u003eO-NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026ndash;\u003c/sup\u003e). Application to the Toulon karst system (Dordogne, France). J Hydrol 610:127972. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jhydrol.2022.127972\u003c/span\u003e\u003cspan address=\"10.1016/j.jhydrol.2022.127972\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMa Y, Jones DL, Wang J, Cardenas LM, Chadwick DR (2021) Relative efficacy and stability of biological and synthetic nitrification inhibitors in a highly nitrifying soil: Evidence of apparent nitrification inhibition by linoleic acid and linolenic acid. Eur J Soil Sci 72:2356\u0026ndash;2371. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/ejss.13096\u003c/span\u003e\u003cspan address=\"10.1111/ejss.13096\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMeng Q, Xing L, Liu L, Xing X, Zhao Z, Zhang F, Li C (2021) Time-lag characteristics of the response of karst springs to precipitation in the northern China. Environ Earth Sci 80:348. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s12665-021-09640-4\u003c/span\u003e\u003cspan address=\"10.1007/s12665-021-09640-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePandit S, Grassian VH (2022) Gas-phase nitrous acid (HONO) is controlled by surface interactions of adsorbed nitrite on common indoor material surfaces. Environ Sci Technol 56:12045\u0026ndash;12054. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1021/acs.est.2c02042\u003c/span\u003e\u003cspan address=\"10.1021/acs.est.2c02042\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchaefer MV, Ying SC, Benner SG, Duan Y, Wang Y, Fendorf S (2016) Aquifer Arsenic Cycling Induced by Seasonal Hydrologic Changes within the Yangtze River Basin. Environ Sci Technol 50:3521\u0026ndash;3529. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1021/acs.est.5b04986\u003c/span\u003e\u003cspan address=\"10.1021/acs.est.5b04986\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSerna A, P\u0026auml;tsch J, D\u0026auml;hnke K, Wiesner MG, Christian Hass H, Zeiler M, Hebbeln D, Emeis K-C (2010) History of anthropogenic nitrogen input to the German Bight/SE North Sea as reflected by nitrogen isotopes in surface sediments, sediment cores and hindcast models. Cont Shelf Res 30:1626\u0026ndash;1638. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.csr.2010.06.010\u003c/span\u003e\u003cspan address=\"10.1016/j.csr.2010.06.010\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShoaib AM, Aly SM, Awad ME, Foo DCY, El-Halwagi MM (2008) A hierarchical approach for the synthesis of batch water network. Comput Chem Eng 32:530\u0026ndash;539. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.compchemeng.2007.03.015\u003c/span\u003e\u003cspan address=\"10.1016/j.compchemeng.2007.03.015\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSiddique AA, Dixit MK, Tembe BL (2016) Molecular dynamics simulations of Ca\u003csup\u003e2+\u003c/sup\u003eCl\u003csup\u003e\u0026ndash;\u003c/sup\u003e ion pair in polar mixtures of acetone and water: Solvation and dynamical studies. Chem Phys Lett 662:306\u0026ndash;316. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.cplett.2016.09.061\u003c/span\u003e\u003cspan address=\"10.1016/j.cplett.2016.09.061\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eŠturm M, Kacjan-Maršić N, Lojen S (2011) Can δ\u003csup\u003e15\u003c/sup\u003eN in lettuce tissues reveal the use of synthetic nitrogen fertiliser in organic production? J Sci Food Agric 91:262\u0026ndash;267. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/jsfa.4179\u003c/span\u003e\u003cspan address=\"10.1002/jsfa.4179\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVadas PA, Owens LB, Sharpley AN (2008) An empirical model for dissolved phosphorus in runoff from surface-applied fertilizers. Agric Ecosyst Environ 127:59\u0026ndash;65. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.agee.2008.03.001\u003c/span\u003e\u003cspan address=\"10.1016/j.agee.2008.03.001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang S, Lin W, Ye Q, Lv W, Liao P, Yu J, Mu C, Wu L, Muneer MA, Zhang Y, Zhan R, Wu A, Wu H, Chen C, Su H, Chen Y, Wang Z, Zhang Z, Ji H, Guo L (2024) Effects of Different Nitrogen Fertilizer Rates on Soil Magnesium Leaching in Tea Garden. J. Soil Sci. Plant Nutr 24:6630\u0026ndash;6640. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s42729-024-01995-4\u003c/span\u003e\u003cspan address=\"10.1007/s42729-024-01995-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWells NS, Maher DT, Erler DV, Hipsey M, Rosentreter JA, Eyre B.D, n.d. Estuaries as Sources and Sinks of N\u003csub\u003e2\u003c/sub\u003eO Across a Land Use Gradient in Subtropical Australia. Glob. Biogeochem. Cycles\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWoodward TR, Frame WH, Alley MM, Whitehurst GB, Whitehurst BM (2011) Design and validation of a laboratory system for measurement of volatilized ammonia. Agron J 103:38\u0026ndash;44. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2134/agronj2010.0242\u003c/span\u003e\u003cspan address=\"10.2134/agronj2010.0242\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu H, Wang Y, Liu Y, Guo S, Zhong Z (2025) Experimental and simulation study of flue gas desulfurization using the limestone-gypsum wet method under an oxygen-enriched combustion atmosphere. J Energy Inst 119:101999. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.joei.2025.101999\u003c/span\u003e\u003cspan address=\"10.1016/j.joei.2025.101999\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu P, Xiao Q, Guo Y, Prelovšek M, Yu Q, Wang Q (2022) Migration, transformation and nitrate source in the Lihu Underground River based on dual stable isotopes of δ\u003csup\u003e15\u003c/sup\u003eN-NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026ndash;\u003c/sup\u003e and δ\u003csup\u003e18\u003c/sup\u003eO-NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026ndash;\u003c/sup\u003e. Environ Sci Pollut Res 29:48661\u0026ndash;48674. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s11356-022-19277-w\u003c/span\u003e\u003cspan address=\"10.1007/s11356-022-19277-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXiong G, Zhu X, Liu M, Yang Y, Chen G, Fu T, Ding R, Xu X, Wu J (2023) Nitrogen cycle pattern variations during seawater-groundwater-river interactions enhance the nitrogen availability in the coastal earth critical zone. J Hydrol 624:129932. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jhydrol.2023.129932\u003c/span\u003e\u003cspan address=\"10.1016/j.jhydrol.2023.129932\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYan J, Chen J, Zhang W (2023) A new probabilistic assessment process for human health risk (HHR) in groundwater with extensive fluoride and nitrate optimized by non parametric estimation method. Water Res 243:120379. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.watres.2023.120379\u003c/span\u003e\u003cspan address=\"10.1016/j.watres.2023.120379\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang Y, Wu J, Du J, Yang, Yunhua, Ye F, Wang Y, Hong Y (2024) Coupled Stable Isotope Tracing and Sulfamic Acid Reduction (SIT\u0026ndash;SAR) Method to Determine the Ammonia and Nitrite Oxidation Rates in Water and Sediments. Anal Chem 96:4197\u0026ndash;4204. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1021/acs.analchem.3c05461\u003c/span\u003e\u003cspan address=\"10.1021/acs.analchem.3c05461\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYi L, Xia R, Tang J, Lu D, Wang Z, Lu H, Zhao L (2023) Modeling the effect of the submerged zone on nitrogen removal efficiency of a bioretention system under dry-wet alterations. J Hydrol 623:129788. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jhydrol.2023.129788\u003c/span\u003e\u003cspan address=\"10.1016/j.jhydrol.2023.129788\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYi L, Xia R, Tang J, Lu D, Wang Z, Lu H, Zhao L (2015) Karst conduit hydro-gradient nonlinear variation feature study: case study of Zhaidi karst underground river. Environ Earth Sci 74:1071\u0026ndash;1078. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s12665-014-3963-5\u003c/span\u003e\u003cspan address=\"10.1007/s12665-014-3963-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYu L, Zheng T, Yuan R, Zheng X, Liu P, Zhang C, Xue C, Mu Y, Liu J, Zhang Y, Tian D, Ye C, Zhang H, Guan J, Iqbal J, Su C, Wang M, Abbas H, Baloch MYJ, Ghani J, Ullah Z, Huq ME (2022) APCS-MLR model: A convenient and fast method for quantitative identification of nitrate pollution sources in groundwater. J Environ Manage 314:115101. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jenvman.2022.115101\u003c/span\u003e\u003cspan address=\"10.1016/j.jenvman.2022.115101\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYu Z-Q, Nakagawa K, Berndtsson R, Hiraoka T, Suzuki Y (2022) Effects of the Japanese Nitrate Directive Plan (NDP) to curb groundwater nitrate-nitrogen content in the Miyakonojo River basin. J Hydrol 615:128563. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jhydrol.2022.128563\u003c/span\u003e\u003cspan address=\"10.1016/j.jhydrol.2022.128563\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYuan Z, Liao Y, Zheng M, Zhuo M, Huang B, Nie X, Wu X, Li D (2020) Relationships of nitrogen losses, phosphorus losses, and sediment under simulated rainfall conditions. J Soil Water Conserv 75:231\u0026ndash;241. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2489/jswc.75.2.231\u003c/span\u003e\u003cspan address=\"10.2489/jswc.75.2.231\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang C, Luo X, Deng Y, Deng Z, Xu R, Amer MA, Ali EAE, Jiang J, Chen H (2024) Insights into enhanced pollutant removal from road runoff by functional microorganisms in a field-scale bioretention facility. J Water Process Eng 62:105294. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jwpe.2024.105294\u003c/span\u003e\u003cspan address=\"10.1016/j.jwpe.2024.105294\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang M, Forte-Perri V, Sun W, Tang L, Huang S, Guo T, Chen X, Li Q (2023) Identification and Observation of Infection Processes of \u003cem\u003eColletotrichum\u003c/em\u003e Species Associated with Persimmon Anthracnose in Guangxi, China. J Geochem Explor 107:1670\u0026ndash;1679. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1094/PDIS-06-22-1372-SR\u003c/span\u003e\u003cspan address=\"10.1094/PDIS-06-22-1372-SR\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang P, Wang X-D, Yue F-J, Chen S-N, Liu Z-H, Lu M, Shi Z-Y (2023) Dynamic characteristics of nitrogen transport in various land use in a typical karst catchment during rainfall events. Sci Data 82:332\u0026ndash;343. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s12665-023-10980-6\u003c/span\u003e\u003cspan address=\"10.1007/s12665-023-10980-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang P, Wang X-D, Yue F-J, Chen S-N, Liu Z-H, Lu M, Shi Z-Y, Olarinoye T, Gleeson T, Marx V, Seeger S, Adinehvand R, Allocca V, Andreo B, Apa\u0026eacute;stegui J, Apolit C, Arfib B, Auler A, Bailly-Comte V, Barber\u0026aacute; JA, Batiot-Guilhe C, Bechtel T, Binet S, Bittner D, Blatnik M, Bolger T, Brunet P, Charlier J-B, Chen Z, Chiogna G, Coxon G, De Vita P, Doummar J, Epting J, Fleury P, Fournier M, Goldscheider N, Gunn J, Guo F, Guyot JL, Howden N, Huggenberger P, Hunt B, Jeannin P-Y, Jiang G, Jones G, Jourde H, Liso IS, Liu Z, Mar\u0026eacute;chal J-C, Massei N, Mazzilli N, Mudarra M, Parise M, Pu J, Ravbar N, Sanchez LH, Santo A, Sauter M, Seidel J-L, Sivelle V, Skoglund R\u0026Oslash;, Stevanovic Z, Wood C, Worthington S, Hartmann A (2020) Global karst springs hydrograph dataset for research and management of the world\u0026rsquo;s fastest-flowing groundwater. Sci Data 7:59 Karmann, I, Koit, O, Kordilla, J, Labat, D, Ladouche, B. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41597-019-0346-5\u003c/span\u003e\u003cspan address=\"10.1038/s41597-019-0346-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang Q, Xu C-Y, Chen YD, Ren L (2011) Comparison of evapotranspiration variations between the Yellow River and pearl river basin, china. Stoch Environ Res Risk Assess 25:139\u0026ndash;150. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00477-010-0428-6\u003c/span\u003e\u003cspan address=\"10.1007/s00477-010-0428-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhao Y, Deng Q, Lin Q, Zeng C, Zhong C, Sullivan PF, Stokes MC, McMillan CK, Weintraub MN (2020) Cadmium source identification in soils and high-risk regions predicted by geographical detector method. Environ Pollut 263:114338. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.envpol.2020.114338\u003c/span\u003e\u003cspan address=\"10.1016/j.envpol.2020.114338\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZuo R, Zheng S, Liu X, Wu G, Wang S, Wang J, Liu J, Huang C, Zhai Y (2023) Groundwater table fluctuation: A driving force affecting nitrogen transformation in nitrate-contaminated groundwater. J Hydrol 621:129606. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jhydrol.2023.129606\u003c/span\u003e\u003cspan address=\"10.1016/j.jhydrol.2023.129606\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang X, Xu S, Zhan J, Ma JW, Liu Junxin L, Lin G Xuesong (2021) Analysis of the effect of two-stage aerobic co-composting of rural organic solid waste. J Environ Eng 15:1708\u0026ndash;1715\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHu QQ (2024) Pollution characteristics and risk assessment of persistent organic pollutants in typical karst industrial areas (D).Guilin University of Technology, China. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.27050/d.cnki.gglgc.2024.000488\u003c/span\u003e\u003cspan address=\"10.27050/d.cnki.gglgc.2024.000488\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXue JY, Liu WT, Jiang XY, Zhao ZH, Zhang L, Cai YG, Wang XL (2019) Ammonia nitrogen cycling rate and influencing factors of water bodies in the middle reaches of the Yangtze River coastal zone. Journal of Yangtze River Basin Resources and Environment 28, 2735\u0026ndash;2742. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/CNKI:SUN:CJLY.0.2019-11-020\u003c/span\u003e\u003cspan address=\"https://doi.org/CNKI:SUN:CJLY.0.2019-11-020\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"environmental-earth-sciences","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"enge","sideBox":"Learn more about [Environmental Earth Sciences](https://www.springer.com/journal/12665)","snPcode":"12665","submissionUrl":"https://submission.nature.com/new-submission/12665/3","title":"Environmental Earth Sciences","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Bayesian isotope mixing model, Nitrate source apportionment, Karst aquifer, Uncertainty quantification, Environmental risk assessment","lastPublishedDoi":"10.21203/rs.3.rs-8680219/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8680219/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eKarst basins are characterized by high permeability and complex surface-subsurface hydrological connectivity which intensifies nitrogen migration. As a typical karst basin located in southern Guilin, the Lipu River basin faces environmental pressures from intensive agriculture and industrial activities; however, the contributions of different nitrate sources and the dominant nitrogen cycling processes have not been systematically characterized. In this study, NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e stable isotopes (δ\u003csup\u003e15\u003c/sup\u003eN-NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e and δ\u003csup\u003e18\u003c/sup\u003eO-NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e, hydrochemical parameters, and a Bayesian isotope mixing model were integrated to identify nitrate sources and elucidate nitrogen cycling in a karst river basin in Lipu County, southwestern China. The results show that total nitrogen concentrations in river water were significantly higher in winter (7.09\u0026thinsp;\u0026plusmn;\u0026thinsp;6.58 mg/L) than in summer (2.95\u0026thinsp;\u0026plusmn;\u0026thinsp;1.48 mg/L). Concentrations of NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e in river water during summer were significantly lower than those in winter and were also lower than groundwater concentrations in both seasons, indicating a strong dilution effect during the rainy season. Among different land-use types, groundwater NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003e concentrations in industrial areas were significantly higher than those in agricultural and residential areas. These spatiotemporal patterns suggest that nitrogen pollution control should prioritize elevated winter nitrogen loads in river water, as well as nitrate contamination in groundwater associated with tributaries and industrial zones. Isotopic signatures combined with modeling indicate that manure and sewage contributed 56.7\u0026thinsp;\u0026plusmn;\u0026thinsp;8.2% of total nitrogen inputs. This integrated isotopic and hydrochemical approach elucidates nitrogen distribution and source contributions in karst basins, providing scientific support for targeted mitigation strategies to reduce nitrogen-related environmental and health risks.\u003c/p\u003e","manuscriptTitle":"Probabilistic source apportionment and quantification of nitrate contamination in a karst aquifer system: Revealed by stable isotopic and hydrochemical proxies","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-13 10:10:55","doi":"10.21203/rs.3.rs-8680219/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-05-16T20:16:42+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-25T02:03:58+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"192231514841470093075984486457046153780","date":"2026-04-25T01:43:31+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"111369334365491294672928952408379369205","date":"2026-04-24T14:14:55+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"201671402823457425440451181880418847302","date":"2026-04-24T08:00:04+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"310124043157938003423192658116308014820","date":"2026-04-23T13:52:46+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-02-10T07:55:15+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-01-27T10:23:32+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-01-27T10:23:05+00:00","index":"","fulltext":""},{"type":"submitted","content":"Environmental Earth Sciences","date":"2026-01-23T14:27:43+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"environmental-earth-sciences","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"enge","sideBox":"Learn more about [Environmental Earth Sciences](https://www.springer.com/journal/12665)","snPcode":"12665","submissionUrl":"https://submission.nature.com/new-submission/12665/3","title":"Environmental Earth Sciences","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"cd74ae66-efd1-41de-a81b-b8dd94fcf191","owner":[],"postedDate":"February 13th, 2026","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"","date":"2026-05-16T20:16:42+00:00","index":34,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-02-13T10:10:55+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-13 10:10:55","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8680219","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8680219","identity":"rs-8680219","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}