Characteristics of leachate from refuse transfer stations in rural China

The properties of leachate from refuse transfer stations (RTSs) in rural China were indefinite. In this study, a total of 14 leachate samples from RTSs in nine provinces of China were characterized for their pH, electric conductivity, chromaticity, concentration of organic substances, nitrogen distribution, volatile organic compounds (VOCs), organic phosphorous pesticide, and heavy metals. The structural composition of fluorescent dissolved organic matter (FDOM) was also determined. To evaluate the leachate pollution potential in this study, a leachate pollution index was derived and used. Chromium (Cr) was the most polluting heavy metal present in rural leachate. Ethanol and ethyl acetate were the most frequently detected VOCs at high concentrations. Three-dimensional fluorescence excitation-emission matrix spectra were used to characterize the FDOM. Three components, tryptophan (C1), tyrosine-like (C2), and humic acid– and fulvic acid–like (C3) substances, were identified from all 14 samples. Tryptophan was the major component of FDOM and present in 45.7% of the samples by calculating the fluorescence intensity percentage, on average. Pearson correlations revealed that the fluorescence intensity of C1 and C3 was strongly related to soluble chemical oxygen demand and dissolved oxygen carbon, while C2 had significant positive correlations with ammonia nitrogen and total phosphorus of the solid waste. This study provided detailed data and findings that could serve as a preliminary basis for broadening options for the treatment and management of leachate from rural RTSs in China.


Introduction
Refuse transfer stations (RTSs) increasingly constitute an integral part of modern solid waste (SW) management systems in meeting the burgeoning challenge that SW disposal poses worldwide. RTSs are light-industrial facilities where SW is uploaded via typical waste collection vehicles and then transported to final disposal sites (e.g., landfills, materials recovery facilities, dumpsites) (Bovea et al. 2007;Kollikkathara et al. 2009;Washburn 2012;Zhen-Shan et al. 2009). When stored in an RTS, SW is usually compressed for its later disposal, which leads to the generation of large amounts of fresh leachate. This leachate is produced via biological and chemical reactions of the waste material (Abunama et al. 2017). Generally, leachate extracted from SW contains high concentrations of many nutrients, but also heavy metals, and even extremely toxic organic compounds such as organic phosphorus pesticides (OPPs), thus posing a serious threat to both the environment and human health (Renou et al. 2008). Hence, developing appropriate treatments for leachate before its discharge is imperative. Recently, mounting studies focused on novel, effective wastewater treatment techniques for leachate have found differential treatment efficiency against leachate dependent on its various components. For example, under a conventional anaerobic digestion process, the optimal nutrient ratio (of carbon, nitrogen, and phosphorus) is biochemical oxygen demand (BOD):N:P = 200:5:1. The chemical oxygen demand (COD) removal rate of fresh leachate can reach 94-96% under optimum operation conditions by using an expanded grantengular sludge bed bioreactor (Liu et al. 2011). Furthermore, pH is also a crucial factor to consider in treatments of RTS leachate: after adjusting the pH in a semi-continuous anaerobic treatment, the COD removal increased from 37.0 to 52.7% (Ghasimi et al. 2008(Ghasimi et al. , 2009. Advanced oxidation processes, such as Fentonlike process, may also be influence by variation in solution pH which could lead to the ineffective decomposition of oxidants (Jung et al. 2017). Notably, the chemical composition of different leachates can vary markedly despite having the same COD value, resulting in an inferior performance of the treatment process (Teng et al. 2021). Accordingly, the systematic characterization of organic and inorganic parameters, heavy metals, and structural characterization of dissolved organic matter (DOM) is crucial to fully understanding the composition of leachate. Such knowledge is needed to provide and discriminate among treatments for their effective use in leachate management.
Several studies have focused on the characterization of landfill leachate (Gu et al. 2022;Wang et al. 2021;Szymanska-Pulikowska 2020, 2021). Yet for RTS leachate, research so far has only investigated seasonal differences in the characteristics of leachate in urban RTSs (Zhao et al. 2013). Barely any studies can be found dedicated to the comprehensive characteristics of leachate from rural RTSs in China. Not surprisingly, lacking this information results in weak or non-existent treatments and disposal regulations for solid waste (Ansari et al. 2018). Arguably then, understanding and quantifying leachate contamination is essential for choosing the proper and effective leachate management and treatment processes to apply. The leachate pollution index (LPI) is a commonly used tool for gauging the polluting potential of leachate from sanitary landfills and open dumpsites (Rajoo et al. 2020). In this study, LPI was first applied to evaluate contamination extent of RTS leachate and then contrasted with published LPI values for other landfills or RTSs.
The objectives of this study were (1) to systematically characterize the fundamental properties of RTS leachate; (2) to identify the composition of DOM; (3) to evaluate the leachate's polluting capacity; and (4) to analyze the relationships between leachate and SW properties. The aim was to provide an empirical preliminary basis for improving the options for treatment and management of RTS leachate in different districts of rural China.

Fresh leachate and solid waste sampling
Leachate samples investigated in this study were obtained from a total of 14 different rural RTSs located in nine provinces of China (Fig. 1), from October to November 2020. In Table 1 are details of each RTS, including its latitude, longitude, capacity, and compression mode. These stations have a daily transferring capacity of approximately 5-210 tons (mostly 5-20 tons) of domestic solid waste. After collecting them, the fresh leachate samples were moved to the laboratory as soon as possible, where they were stored in a freezer at − 20 °C to minimize alterations to their biological and chemical properties. It should be noted no rain fell during the days when these samples were collected.
At the same time, approximately 20 kg of each solid waste (SW) samples was collected according to method of coning and quartering (Ajay et al. 2022), and stored at − 20 °C until their particle size distribution analysis and component characterization. Sieves with different apertures were used, manually, to analyze the particle size distribution. Considering the complexity and irregularity of SW, the sieving process was separated into two parts. For hard matter, its particle size was measured according to the longest side; for soft matter (i.e., plastics, textiles, and papers), particle size was measured according to the sieve's mesh through which the same dropped waste passed, in triplicate, under its natural coiled state, as depicted in Fig. S1. In this study, the particle size was classified into five categories: < 50 mm, 50-100 mm, 100-150 mm, 150-250 mm, and > 250 mm.

Analytical methods
The leachate samples' pH was measured with a pH meter (S470 Seven Excellence, Mettler Toledo, Shanghai, China). Their chromaticity was quantified using a multiparameter water quality analyzer (GL-900, Gelin Kerry Precision Instrument Co., Ltd., Shandong, China), while the electric conductivity (EC) of leachate was recorded with a conductivity meter (DDS-307A model, Shanghai Precision & Scientific Instrument, Shanghai, China). Soluble and total chemical oxygen demand (respectively SCOD and COD) was measured by the fast digestionspectrophotometric method (HJ/T 399-2007). To measure the NH 3 -N content, Nessler's reagent spectrophotometry was used (HJ 535-2009). For the total nitrogen (TN) content, it was determined by applying the alkaline potassium persulfate digestion-UV spectrophotometric method (HJ 636-2012). DOC and dissolved inorganic carbon were determined by a TOC analyzer (Aurora 1030C, O.I. Analytical Co., Ltd., USA) with an autosampler (Model 1088, O.I. Analytical Co., Ltd., USA). The concentrations of copper (Cu), nickel (Ni), chromium (Cr), and zinc (Zn) were detected and quantified using ICP-OES (iCAP 7200 ICP-OES, Thermo Fisher, USA). The concentrations of arsenic (As) and total mercury (THg) were determined by an atomic fluorescence spectrometer (Haiguang AFS-8510, Beijing, China). The concentrations of lead (Pb)  For the excitation-emission matrix (EEM) analysis, each leachate sample was filtered through a 0.45-μm membrane and then diluted 100 times with ultrapure water (18 MΩ·cm). The fluorescence EEM was measured by an Aqualon fluorescence spectrometer (HORIBA Instruments Inc., Irvine, CA, USA), as described in our previous study (Su et al. 2021). The Aqualog-coupled software was equipped with a built-in tool for the normalization of water Raman scattering as well as the correction of the IFE and Rayleigh masking effects (Quatela et al. 2018). Normalization of water Raman scattering was used to remove the instrument-dependent factors; this was done by dividing the measured intensities by the Raman peak intensity of ultrapure water at an Ex = 350 nm. Once standardized, EEM data were further analyzed using PARAFAC in the Solo + MIA 8.6.1 software (Eigenvector Research, Manson, WA, USA), with a non-negativity constraint applied to the parameters (Hunt and Ohno 2007). The model outputs for the EEM-PARAFAC components were then compared with matched emission and excitation spectra available in the OpenFluor database (https:// openfl uor. labli cate. com/) (Murphy et al. 2014). Pearson correlations were implemented in SPSS 25 software (IBM, USA). The program Origin 2021 was used to analyze the basic statistical parameters (i.e., max., min., mean, standard deviation [SD], and outliers).

Results and discussion
The diagrams in the following sections summarize the chief descriptive statistical terms used, with respect to (1) central tendency in the data, expressed using the mean [average] and median, and (2) data variability, in terms of the range (min.-max.), first quartile-third quartile (Q1-Q3), interquartile range (IQR), and outlier.

Leachate pH, conductivity, and chromaticity
The pH values of leachate samples from the different RTSs are shown in Fig. 2 a. As depicted, the pH ranged from 3.69 to 8.72, with a Q1-Q3 of 4.25-7.3, and a mean of 5.98; hence, the leachate had acidic to alkalescency characteristics. Unlike the variation in rural leachate, typical urban leachate is predominately acidic. One study found that the pH of leachate from an RTS in Shanghai ranged from 3.65 to 4.99 and 2.69 to 4.47 during the winter and summer (Zhao et al. 2013). The relative higher pH of rural leachate is probably due to its lower proportion of kitchen garbage vis-à-vis urban leachate. A greater accumulation of volatile acids enables hydrolytic and acidogenic bacteria to become dominate, finally leading to reductions in pH (Sun et al. 2011). Geographically, in our study, all 5 samples with a pH > 7 were collected from northern China (Z1, Z2, B2, Z3, and P1), implying that the lower temperatures there led to higher pH levels. This phenomenon was also consistent with a previous study which reported that higher temperatures may cause a lowering of leachate pH (Rafizul and Alamgir 2012), due to more accumulation of volatile acids. Therefore, seasons, climate, and temperature change can all influence the pH of RTS leachate. In Fig. 2 b are the chromaticity results for leachate from different rural RTSs. Chromaticity is often used to characterize the color of dissolved substances in water. The chromaticity of leachate samples ranged from 2100 to 26,700; however, their Q1-Q3 was 2888-10,612. Only three samples (from B1, P1, and Z3) had a chromaticity above 11,000. The high chromaticity of leachate might arise from a longer retention time and more humic acid-and fulvic acid-like substances produced by the degradation of SW by microorganisms.
As seen in Fig. 2 c, the EC of RTS leachate ranged from 2.71 to 22.85 ms/cm. The values for Q1-Q3 were 9.17-18.33 ms/cm. Rural RTS leachate's EC is slightly higher than what is reported for urban leachate; for example, the EC of leachate in Shanghai RTS ranged from 4.41 to 13.82 ms/cm (Zhao et al. 2013). A higher EC value could arise from more organic acids formed during macromolecular organic compounds' biodegradation process (Sivula et al. 2012). The dilution of leachate on rainy day can lower EC values (Yuen et al. 2001).

SCOD and DOC of leachate
COD and TOC concentrations are usually used to indicate the amount of organic substances in water. Here, to avoid interference from solid particles, all leachate samples were filtered through a 0.45-μm membrane before their analysis for SCOD and DOC. Figure 3 a depicts the SCOD and DOC concentrations' distribution across different rural RTSs of China. Excluding for two outliers (defined as < Q1 -1.5IQR or > Q3 + 1.5IQR), these corresponding to samples X2 (SCOD = 66 100 mg/L, DOC = 22 167.5 mg/L) and X3 (SCOD = 62 500 mg/L, DOC = 15 600 mg/L), the SCOD ranged from 2900 to 28,700 mg/L while the DOC ranged from 801 to 9313.5 mg/L. For landfill leachate, there is close relationship between COD/TOC and the duration of its decomposition process (Wdowczyk and Szymanska-Pulikowska 2021). For urban RTS leachate, however, temperature is perhaps the most pivotal factor: previous study showed that both COD and TOC in winter were much higher than those in summer (Zhao et al. 2013). Further, COD and TOC concentrations of leachate in urban RTSs can reach 32,640-157,200 mg/L and 150,000-45,000 mg/L, amounts much higher than those of rural leachate in our study. Daily output, rainfall dilution, and self-micro-degradation under different seasonal conditions are the main factors driving variation in COD and TOC concentrations of leachate (Christensen et al. 2001;Lou et al. 2011).

TN and NH 3 -N of leachate
Quantifying the TN and NH 3 -N content is essential when characterizing the pollution level of leachate. Figure 3 b presents the TN and NH 3 -N concentrations of rural RTS leachate. Across the nine rural provinces, its TN and NH 3 -N content respectively ranged from 125 to 1669.5 mg/L and 114.85 to 1174 mg/L, indicating that NH 3 -N accounted for the most of TN. Similar to SCOD and DOC, the TN of rural leachate was much lower than that of urban leachate. Zhao et al. (2013) found that TN concentration of RTS leachate in Shanghai reached 697 to 5332 mg/L in winter. On the contrary, the NH 3 -N concentration of that city's urban leachate was only 39 to 568 mg/L, and thus lower than rural leachate, this likely explained by the dominant proportion of NO 3 − -N in urban leachate. Unlike urban leachate, the samples of rural leachate in our study may have been stored for hours to days before they were collected from storage tanks. Organic nitrogen such as fat and protein might be converted into NH 3 -N via the ammonia reaction under anaerobic conditions (Romain et al. 2008;Yang et al. 1999), which could

Fluorescence spectra-parallel factor components of FDOM
FDOM has a strong absorption of ultraviolet and visible light, which leads to the dark color of leachate (Lozinski et al. 2019). The composition of FDOM is crucial to how well a given leachate treatment process performs. Therefore, the characterization of FDOM was carried out here using typical three-dimensional fluorescence EEM spectra. Fluorescence intensities of 14 leachate samples are depicted in Fig. S2. The calculation of parallel factor components was conducted using PARAFAC, with 2-10 components considered. The optimal number of components was designated and validated by applying core consistency diagnostics followed by a split-half analysis. As Fig. S3 shows, a split-half reliability of 95.9% and core consistency score of 93% were obtained, and the model explained 98% of the variation when the component number was set as 3; hence, three effective components were identified within the fluorescence EEM spectra in this study. The representative EEM and fluorescence spectral loadings for these three common components (here labeled C 1 , C 2 , and C 3 ) appear in Fig. 4. By comparing the three components' excitation and emission loading data with the OpenFluor database (www. openf luor. org), a group of previously identified similar components was acquired (Table S1).
The decomposed C 1 and C 2 were identified as characteristic peaks at an Ex/Em of 276/340 nm and 270/301 nm, respectively. Component C 3 featured a primary excitation maxima at approximately 330 nm followed by a lower intensity secondary peak at 246 nm and an emission maxima at 422 nm. According to the search results of the Openfluor database, C 1 was the ubiquitous "protein" peak bearing the spectral features of the amino acid tryptophan, but there was also a probable presence of lignin-derived phenols generated from rotting woods (Coble 2007;Cohen et al. 2014;Jia et al. 2017;Lee et al. 2018;Osburn et al. 2012;Yamashita et al. 2011), while C 2 was characterized as tyrosine-and proteinlike substances (Chen et al. 2018;Smith et al. 2021), with C 3 instead corresponding to humic/fulvic acid-like substances in the leachate (Dainard and Gueguen 2013).
The maximum fluorescence intensity (F max ) overall and of each component are described in Fig. S4. Evidently, regarding the F max of sample C1, the values for samples X2 and X3 surpassed those in the other samples, whereas sample B2 had the lowest fluorescence intensity for all three components. To further investigate the proportion of each component present in each leachate sample, Fig. 5 depicts the relative distributions of the three components calculated by PARAFAC modeling of FDOM. Clearly, component C 1 constituted the largest proportion, averaging 45.7%, while the C 2 and C 3 occurred in mean proportions of 31.9% and 22.4%, respectively. Thus, protein-like substances and humic/fulvic acid were inferred as being the two dominant components of FDOM in China's rural leachate. As a comparison, leachate DOM characterization in landfills varies from different ages, protein-like substance occupies the predominant moieties in young landfill leachate, while humic acid-like and fulvic acid-like substances originating from the condensation and polymerization of microbial degradation by-products are dominant in young landfill leachate (Huo et al. 2008). Protein-like substance, as biodegradable factions, is composed of carbon chained structures and can be simply removed by biotreatment processes while humic acid-like and fulvic acid-like substances are classified as non-biodegradable fractions.

Heavy metals in leachate
Generally, the major sources of heavy metals in SW are batteries, consumer electronics, ceramic products, light bulbs, house dust and paint chips, and lead foils such as used for wine bottle closures, as well as plastics and used motor oils, in addition to some inks and glass material (Abunama et al. 2021). Industrial activities-power plants, mining activities, metal-smelting industry, and chemical plants-are also pivotal sources of heavy metal pollution (Du et al. 2015;Zheng et al. 2010). Due to their stability, heavy metals gradually accumulate in landfills and soil. Given the migration and percolation of toxic heavy metals into groundwater, they pose potentially serious pollution risks that threaten the environment and human health (Adamcova et al. 2017). In Malaysia, it was found that As, Cr, and Pb are the dominant heavy metal contaminants in landfills and surface soils (Hussein et al. 2021). However, generally little attention has been paid to investigating RTS leachate for heavy metal pollution.
As conveyed in Table 2, eight heavy metals (Cu, Zn, Ni, Cr, As, Pb, THg, and Cd) were detected in the leachates of the 14 samples examined in this study, but THg was not detected in every sample. Evidently, pollution levels of Cr were the highest among eight assessed heavy metals, for which 10 of the 14 samples exceed that element's discharge standard value. The Cr and As concentrations in fresh leachate sampled from open-dumping sites in Malaysia also surpassed the leachate discharge standard value (Hussein et al. 2021). Excess amounts of toxic heavy metals, such as As and Cr, are harmful to many environmental organisms. Thus, additional treatments targeting such heavy metals, or proper liners and collection systems during leachate disposal, are necessary in rural areas of China.
Higher contamination levels were found in the leachate of B1, C1, P1, P2, and Z2 than the other nine samples, indicating the occurrence of heavy metals in excess is not spatially random and limited to particular rural areas. Due to disparities in economic improvement of rural areas, especially the penetration of urban consumption and behavior into villages and small towns, heavy metal pollution from electronic wastes has become more pronounced in certain areas. To reduce their heavy metal pollution of rural SW, further management interventions, including electronic waste recycling, reuse, or safe disposal, are indispensable.

VOCs and organic phosphorous pesticides in leachate
Both VOCs and OPPs are common in agricultural and industrial production; excess organic pollutants can generate odor compounds and influence the quality of surrounding air, and also pose a threat to the environmental organisms (Battaglin et al. 2020;Cheng et al. 2020;Liu et al. 2016). However, few studies have reported on their occurrence in rural RTS leachate. To comprehensively investigate the presence and concentration of toxic organic compounds, 109 of them, consisting of 86 VOCs, 17 OCPs, and 6 OPPs, were tested in this study (listed in Table S2). For 13 detected VOCs, their concentrations are presented in Fig. 6 and Table 3. The more harmful OCPs and OPPs were not detected in the rural leachate samples. Importantly, ethanol and ethyl acetate were the most frequently detected organic substances, and the former's average concentration was significantly higher than that of other organics. These results we may attribute to the fermentation process of raw starch materials (e.g., cereals, potatoes, corn, sorghum, or wild plant fruits) and raw sugary materials (e.g., molasses, sulfite waste liquor) under anaerobic conditions. Some of this high-concentration ethanol could have been further esterified with acetic acid in food waste to form ethyl acetate, thereby leading to widespread generation of ethanol and ethyl acetate in rural leachate. It is proved that alcohol, volatile fatty acids, and aldehydes and ketones are produced during carbohydrate degradation (Fang et al. 2012;Scaglia et al. 2011).
Furthermore, according to the list of carcinogens published by World Health Organization's International Agency for Research on Cancer, some of the other detected organic compounds are potentially carcinogenic. Among them, the xylenes, toluene, and methyl tert-butyl ether are classified as group III carcinogens. Exposure to toluene may lead to central nervous dysfunction, vascular dilation, and erythema, as well as liver, kidney, and heart injury (Kim et al. 2019). Naphthalene, ethylbenzene, and 1,2-dichloromethane are classified as group II carcinogens. Due to their toxicity and possible carcinogenicity, these VOCs warrant special consideration and due diligence in monitoring. For X3, its concentration of o-xylene, para/m-xylene, and ethylbenzene respectively reached 954, 1030, and 337 μg/L, being far higher than those of other samples. As a comparison, a recent study analyzed the species and concentrations of VOC emissions at a rural RTS in west northern China. It is found that the concentration of N-methylpyrrolidone, methyl isobutyl ketone, dimethyl sulfoxide, chlorobenzene, toluene,  allyl chloride, ether, and ethyl lactate were relatively higher and ranged from 383.14 to 6313.07 μg/m 3 (Chai et al. 2022). Therefore, further investigation and targeted treatment of VOCs should be carried out in this specific RTS.

Leachate pollution index calculation
LPI is a tool that allows the obtained results to be standardized and compared with other leachate in landfills or RTSs (Naveen et al. 2017). To date, most studies have focused on deriving the LPI of landfill leachate. However, LPI's calculation and evaluation for leachate of RTS can also benefit the design of suitable treatment processes. LPI's calculation can be divided into four parts: selecting the parameters, deriving their weights (w i ), formulating the sub-indices (p i ) curves/equations, and choosing the optimum aggregation formula. Here, the LPI formulation basically consisted of 18 parameters, including LPI inorganic parameters (LPI in ): pH, total dissolved solids, Cl, total Kjeldahl nitrogen, and NH 3 -N; LPI organic parameters (LPI or ): biological oxygen demand (BOD), chemical oxygen demand (COD), phenolic compounds, total coliform bacteria, and cyanide; and LPI heavy metal parameters (LPI hm ) Fe, Cu, Ni, Zn, Pb, Cr, Hg, and As. Furthermore, the assigned weight (w i ) to each LPI parameter was set up according to the significance of pollutants, to indicate the relative contribution to the overall polluting potential of a leachate sample. Additionally, the curves of sub-index (p i ) were determined by calculating the sub-index formulation of Kumar and Alappat (Kumar and Alappat 2005) Accordingly, the LPI can be calculated according to Eqs. (1) and (2), as follows: where n and m denote the number of pollutant parameters of the leachate sample; w i is the weight of a given pollutant parameter (all weightings can be found in Table S3) (Kumar and Alappat Babu 2004); and p i refers to the sub-index values of each leachate pollutant variable.
In this study, LPI was calculated using Eq.
(2) because some pollutant parameters were missing. Figure 7 shows that the LPI values of different RTS leachate samples ranged from 15.68 to 42.34, with a corresponding Q1-Q3 of 22.31 to 36.86 and an average value of 27.51. The last is a bit higher than the overall LPI average reported for global landfills and dumpsites, at 22.23 and 15.66, respectively. Their lower pollutant potential compared with the RTSs of rural  Table 3 VOCs and organic phosphorous pesticide concentration in 14 leachate samples China could result from various leachate-dilution factors of landfills and dumpsites, including the age of their waste, climatic conditions, and rainfall regime and levels (Gomez et al. 2019).

Correlation analysis
Correlation analysis is a preliminary descriptive technique to detect and quantify the degree of association between various parameters of interest, which could point to a causal relationship (Mor et al. 2006). Correlations among characteristics of leachate, SW properties, and details of RTSs were conducted in this study. Selected parameters comprised two parts. The first was a leachate index including seven fundamental properties, seven heavy metals, fluorescence intensity (FI) of components (C 1 , C 2 , C 3 , and total FI), and LPI values. The second was a SW index that included total phosphorus (TP), volumetric weight, particle size, and composition of SW. Other parameters investigated were temperature, latitude, and longitude of the sampling locations ( Fig. 1), and scale of RTS compression. Extremely significant positive correlations were observed between SCOD and COD (r = 0.974) and between AN and TN (r = 0.929), as expected, given that these two pairs of variables indicate the organic matter and nitrogen levels of leachate. We found pH negatively correlated with other fundamental parameters including conductivity, SCOD, COD, and TN, whereas EC was significantly and positively correlated with COD, AN, TN, heavy metals, and FI.
Obvious positive correlations were observed between various heavy metals, indicating their similar chemical source and behavior. For example, the Pearson correlation coefficients r Zn-Cu , r Cu-Cr , r Zn-Cr , and r Pb-Cd were 0.944, 0.931, 0.931, and 0.843, respectively. Similar results were also found in a global scale investigation (Abunama et al. 2021). The close relationships among Cu, Zn, and Cr may imply the presence of industrial electroplating sewage in China's rural leachate (Chen et al. 2020;Huang et al. 2022). Traffic emissions including tire abrasion, lubricants, and the corrosion of vehicular parts may also cause Cu and Zn pollution (Atiemo et al. 2011). Additionally, Pb and Cd had positive correlation. Coal combustion might be responsible for the Pb and Cd pollutants in leachate due to the high consumption of coal in most rural areas of China. According to one report (Tian et al. 2012), the Cd and Pb emissions from coal combustion increased from 31.14 and 2671.73 tons in 1980 to 261.52 tons and 12 561.77 tons in 2008, respectively (Fig. 8).
FI showed a significant positive correlation with SCOD (r = 0.862), COD (r = 0.822), AN (r = 0.779), and TN (r = 0.0.745). Nevertheless, three components C 1 , C 2 , and C3 were each positively correlated with different parameters. C 1 and C 3 had stronger correlations with SCOD and COD (r C1-SCOD = 0.914, r C1-COD = 0.911, r C3-SCOD = 0.896, r C3-COD = 0.847), while C2 was better correlated with AN (r = 0.802) and TP (r = 0.749), suggesting this component had higher nutrient content and harbored a protein-like substance. The proportion of organophosphorus significantly increased in SWTP, likely because inorganic phosphorus is sourced primarily from laundry detergent and phosphate fertilizer (among others); its typically high solubility means that it can easily be carried away in surface runoff into the wastewater treatment system. Kitchen food waste is the major source of organic compounds in rural leachate we sampled, while protein is the main ingredient of food waste as well. Collectively, this would reasonably explain the abundance of protein in leachate.
For LPI, Pearson correlation coefficients of LPI and LPI IN were r = 0.929 (0.8 < r < 1.0) and P = 0.000 (i.e., P < 0.01), indicating that the LPI value was highly dependent on LPI IN . Although the contamination by heavy metals cannot be ignored given the conclusion drawn in subsection 3.4, it had minimal impact on LPI HM (Vaccari et al. 2019).
Based on the review by Abunama et al. (2021), who summarized the leachate data from landfills and dumpsites in countries worldwide, the LPI rating is closely related to the age of waste disposal facilities: those < 5 years old have a high LPI average = 26.5; those 5-10 years old have an intermediate LPI average = 23.5, and those > 10 years old have a low LPI average = 17.5. As shown in Table 4 investigate other factors influencing the LPI value of RTS, the correlations between the composition and particle size distribution of corresponding solid wastes were performed (Fig. S5). We found that the proportion of kitchen garbage had a significant positive correlation with particle size < 50 (r = 0.863) but a negative correlation with paper's proportion (r = − 0.803). Yet no obvious correlation was detectable between particle size, the composition of SW, and LPI value. Accompanying the rapid economic development and improved income levels of rural areas in China, urban consumption patterns have been increasingly adopted in villages and towns. As seen in Fig. S6, although the kitchen waste's proportion reached 45%, still the largest component of rural waste, this remained lower than its proportion of urban waste. This is because food waste in rural areas can be consumed, in part, by feeding it to livestock and poultry, and additionally, leaving higher proportions of paper, rubber, and plastic in the SW that ends up at rural RTSs. More complex and changeable components of SW and leachate might be thus responsible for the low correlations found between LPI and waste properties in rural leachate.
Finally, to further investigate the relationships between leachate, SW properties, and geographical location, the latitude, longitude, and temperature of sampled RTSs were also included in the correlation analysis. Interestingly, among all parameters, concentration of seven detected heavy metals showed a negative correlation with latitude, perhaps because of the greater industry and coal combustion in northern China.

Conclusion
In this study, the leachate from rural refuse transfer stations (RTSs) of China was characterized for its pH, electric conductivity, and chromaticity, in addition to its concentrations of organic substances, nitrogen distribution, VOCs, OPP, and heavy metals. The main conclusions are as follows:  (1) The pH value of rural leachate ranged from 3.69 to 8.72, and showed acidic to alkalescency; its chromaticity ranged from 2100 to 26,700; and its EC ranged from 2.71 to 22.85 ms/cm. The leachate content of SCOD and COD was 2900-28,700 mg/L and 801-9313.5 mg/L, respectively, while TN and NH 3 -N ranged respectively from 125 to 1669.5 and 114.85 to 1174 mg/L. (2) The composition of FDOM was also structurally characterized: three components C 1 , C 2 , and C 3 were identified as tryptophan, tyrosine-like, and humic/fulvic acid-like substances, respectively, whose proportions were 45.7%, 31.9%, and 22.4%, on average of 14 samples. The fluorescence intensity of C 1 and C 3 is strongly correlated with SCOD and DOC, while C2 had significant positive correlations with both AN and TP. (3) Among heavy metals, Cr is the predominant pollutant occurring in rural leachate. Nonetheless, there were clear positive correlations between various heavy metals, e.g., for Zn/Cu/Cr and Pb/Cd. Ethanol and ethyl acetate were the most frequently detected VOCs with high concentration. (4) Finally, the average LPI value in rural China's RTSs was 27.51, this being higher than the overall LPI average for global landfills and dumpsites.
The characteristic of leachate in rural China provides potential useful reference for RTS treatment and SW management. And the investigation on leachate characterization for specific treatment process in RTS is recommended in the future.