Influence of Soil Organic Carbon, Water Holding Capacity, and Moisture Content on Heavy Metals in Rice Paddy Soils of Western Ghats of India

Analysis of soil samples collected from 16 rice paddy fields located in the Western Ghats region was performed to quantify the concentration of Cu, Zn, Mn, Fe, Ni, Cr, Cd, and Pb using atomic absorption spectroscopy. High concentrations of these heavy metals were found in rice paddy fields regularly cultivated using agrochemicals. We compared this concentration with soils of rice paddy field that was not under cultivation. Cu, Zn, Mn, Fe, Cr, Ni, Pb, and Cd showed increases of 1.2, 1.3, 2.3, 2.2, 1.8, 2.8, 1.8, and 8.5 times, respectively, in the rice paddy fields cultivated with synthetic fertilizers such as NPK, urea, potash, diammonium phosphate, etc., and several categories of pesticides belonging to the class organophosphates, carbamates, and acetanilide herbicide. In contaminated sites, the heavy metals exhibited maximum correlation with soil moisture content (SMC) (Zn, Fe, Cr, Ni, and Cd), soil organic content (SOC) (Fe, Cr, Ni, and Cd), and water holding capacity (WHC) (Cu, Pb, and Cd) than those observed for the reference site. The principal component analysis (PCA) revealed a total of 77.944% variance of heavy metals contributed from WHC (40.259%), SMC (20.854%), and SOC (16.832%). This indicates the build-up of heavy metals in rice paddy soils under the strong influence of moisture content, water holding capacity, and organic carbon content of the soil.


Introduction
Heavy metal contamination in agricultural soils have attracted intensive attention in India. Many documented cases are available on heavy metal contamination in agriculture soil due to mining, industry, foundries, power plants, waste water irrigation, and agriculture (Chabukdhara et al., 2016;Dhaliwal et al., 2021;Giri et al., 2017;Kumar & Maiti, 2015;Mishra et al., 2009;Reddy et al., 2013;Sharma et al., 2018;Vasudhevan et al., 2022;Yadav et al., 2017). Such anthropogenic sources contribute external heavy metals to the agricultural soils causing enrichment of metal concentration (Song et al., 2018) and affect the physiological and biological processes of plant (Alloway, 2013;Nagajyoti et al., 2010). As heavy metals are persistent and potentially toxic, metal enrichment of agricultural soils rises concerns over food safety and human health (Ali et al., 2019;Huang et al., 2021;Škrbić et al., 2021). Farming practices are known to influence the heavy metal content of the soils and food produce (Atafar et al., 1 3 Vol:. (1234567890) 2010; Bai et al., 2010;Du et al., 2020;Kuppusamy et al., 2018;Marrugo-Negrete et al., 2017;Satpathy et al., 2014;Shan et al., 2013;Škrbić et al., 2021). Application of agrochemicals, viz, synthetic fertilizers and pesticides contribute unwanted load of heavy metals to the agricultural soils (Reddy et al., 2013). This depending on the soil characteristics and condition either accumulate in the soil or available for plant to uptake (Bhatti et al., 2016;Mishra et al., 2009). Soil characteristics such as pH, organic carbon content, cation exchange capacity, and moisture content controls the metal speciation, mobility, and availability of heavy metal in the soil (Dube et al., 2001;Salem et al., 2020;Zeng et al., 2011). Metal speciation and availability is also influenced by cultivation practices. Studies have revealed that long-term fertilization alters the soil physical and chemical properties, which affects metal distribution and mobility in the soil (Lambert et al., 2007;Liu et al., 2007;Rutkowska et al., 2014). Furthermore, ploughing and puddling of soils found to alter electrochemical, chemical, and microbial processes and influence on metal availability to plant (Witt & Haefele, 2005).
In case of rice paddy soils water saturation is the dominant factor that governs soil development, soil processes, and biotic communities in the soil (Witt & Haefele, 2005). To supplement adequate nutrients, varieties of synthetic fertilizers are used during rice paddy cultivation. Rice paddy is one of the important crops of India cultivated with modern technology and the application of various agrochemicals (Reddy et al., 2013). Agrochemicals, viz., synthetic fertilizers such as NPK, urea, potash, diammonium phosphate, and muriate of potash, and several categories of pesticides belonging to the class organophosphates (viz., phorate, malathion, methyl parathion, monocrotophos etc.,) carbamates (e.g., carbofuran) and pyrethroids (e.g., cypermethrin), nicotinoids, and nereistoxin analogue insecticide are used for rice paddy cultivation in India.
In Western Ghats region of Karnataka, rice paddy is the staple crop and based on availability of water intensive cultivation is practiced. For the better yield and protection of the crop, profuse application of synthetic fertilizers and pesticides are being practiced. Although risk assessment of iron ore mining in surrounding areas including rice fields in Goa (Daripa et al., 2022) and Pondicherry (Satpathy et al., 2014) in southern India is available, information regarding the heavy metal content in rice paddy soils of hill terrains of Western Ghats is not well documented. Therefore, the overall goals of this study were (1) to quantify the selected heavy metals in these rice paddy fields cultivated with and without the use of agrochemicals and (2) to record the accumulation of heavy metals in the surface soils of rice paddy fields and to analyse the possible reasons.

Study Area
The study area is located in a mosaic of different crops of cultivation in the hill terrain of Western Ghats (Fig. 1). The area has a subtropical climate, and the annual precipitation ranges between 769 and 1042 mm. Rice paddy cultivation is being practiced for more than 30 years; however, nearly from the past 10 years, agrochemical application has become a necessary managerial practice. The general information of the site, i.e., cultivation history, cultivation pattern, frequency, and type of agrochemical used in the sites, were collected through a questionnaire. The survey revealed that the current general managerial practice of farming comprises the use of varieties of agrochemicals: synthetic fertilizers, pesticides, biosolids, etc. Most of the farmers in the area have used NPK (nitro-phosphorous and potash 17:17:17 or 20:20:20) or ammonium phosphate and sulfate (20:20:0:13) or ammonium phosphate (16:20:0) or DAP (diammonium phosphate) or urea and potash fertilizers and machete (butachlor, 50% EC, herbicide), chlorpyrifos 20% EC, and dithiocarbamate fungicides ( Table 1). The agrochemical usage varied depending on the managerial practices. In most of these rice paddy fields, farmers used bio-organic fertilizers and biosolids as initial nutrients at the time of tilling of the soil for paddy cultivation (personal observation). Rice paddy cultivation relayed on rainfed irrigation systems. Single cropping per annum with single tillage is the general practice followed. The approximate yield of paddy fields in the area is two tons of grains per acre.

Sampling Site and Soil Sampling
Sampling was performed in the month of February 2018, which was a non-cultivation period in the selected 1 3 Vol.: (0123456789) sites that is generally characterized by low soil moisture content (SMC) (range 1.27-13.00%). Soil samples were collected from 16 different rice paddy fields ( Fig. 1) located between 13°43′49.3″-13°46′14.6″N and 75°38′004″-75°38′29.8″E at altitudes ranging between 622 and 637 m msl. Out of 16 sampling sites, at 15 sites, rice paddies were cultivated with prominent application of agrochemicals and are considered "contaminated" sites, and one site that has not been cultivated with any synthetic agrochemicals is considered a "reference site." Each site has an area of one acre. At each site, soil samples were collected at four locations (so that, 15 × 4 = 60 samples from contaminated sites, and 1 × 4 = 4 samples from reference site) following the methods described by Hoshmand (2006). At the time of soil sampling, the soil temperature (°C) was recorded using a mercury bulb thermometer (precision 0.1 °C). The samples were collected in a ziplock polythene bag. Following each sampling, the soil samples were shade dried, powdered, and sieved using a soil sieve (Mesh size 2 mm -HACH soil testing Kit) and then used for chemical analyses.

Analyses of Soil Samples
The soil organic carbon (SOC %) content was estimated following the Walkley and Black rapid titration method (Nelson & Sommers, 1996). Soil pH, soil moisture content (SMC %), and soil water holding capacity (WHC %) were recorded following the methods of Allen (1989). Following acid extraction (EPA 3050B SW-846), the total heavy metal concentrations (ppm) in the soil samples were analyzed using atomic absorption spectrometry (PerkinElmer PinAAcle 900F) (USEPA, 1996; Ashaiekh et al., 2019).

Statistical Analysis
Statistical data analysis was performed using SPSS V. 20 statistical software and MS Excel (Microsoft 2019). Basic statistical parameters, such as range, mean (x) , median, standard deviation (SD), standard error (SE), and skewness, were computed. Shapiro-Wilk Lambda (w) was employed to test the normality of all data on soil properties and heavy metal concentrations. One-way ANOVA was used to detect the significant differences in the concentrations of various parameters between reference and contaminated soils. Pearson's correlation coefficient was used to test the association between soil parameters and heavy metal concentrations in the soil. Principal component analysis (PCA) is a multivariate statistical tool useful for data evaluation, dimension reduction, and pattern recognition (Škrbić et al., 2005, 2010). In the present study, PCA was performed to determine the dominant factors influencing heavy metal accumulation in the soil. Prior to PCA, the values of soil parameters and heavy metals were transformed to z-scores. The Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy significance < 0.05 is considered an ideal condition for factor reduction.

Parameters of Soil and Heavy Metals in the Reference Site
At the reference site (Loc. 75°38′24.5″N-13°44′25. 4″E, altitude: 627 m msl), the soil temperature ranged between 28 and 30 °C. The pH of the reference site remained acidic to circumneutral, recorded between 5.3 and 7.2. Statistics of the parameters of the soil and heavy metal concentrations of the reference site are presented in Table 2. The SMC and SOC of the reference site were low, while the WHC was recorded up to 25%. The SMC, SOC, and WHC showed a little variance over the study period. At the reference site, concentrations (ppm) of copper (Cu), zinc (Zn), manganese (Mn), iron (Fe), chromium (Cr), nickel (Ni), lead (Pb), and cadmium (Cd) were recorded. The statistics of these parameters indicate that all these values are normally distributed. Furthermore, Table 2 also presents the concentrations of heavy metals recorded at the reference site, with general heavy metal concentrations recorded for typical uncontaminated agricultural soil (Nagajyoti et al., 2010). Except for the concentration of Cu, all other heavy metals recorded in the reference site fall within the upper limit of the concentrations detailed for uncontaminated soils. However, the Cu concentration (140.33 ± 0.021 ppm) was high compared to the range given for uncontaminated sites (Nagajyoti et al., 2010). Table 3 represents the correlation coefficients between the parameters of the soil and heavy metal concentrations at the reference site. The analysis indicated a reduction in the concentrations of Mn, Ni, and Pb with an increase in SMC (%). In contrast, the concentration of Zn showed an increase with SMC at a significant level. The Mn and Ni are also decreased with the increase in SOC. The Cu concentration increased with increasing WHC, while the Cd concentration decreased significantly. Among the correlations of heavy metals, the increase in Cd concentration decreased the Cu content. An increase in lead content concurrently increased the concentrations of Mn and Ni and decreased the Zn concentration. An increase in Fe content showed an increase in Cr content, while a similar increase was observed between Pb and Ni. Among the parameters of the soil, the SMC comprised the maximum correlation with heavy metals. This exhibited R 2 = 0.942; F = 32.667, p = 0.029. The pH values did not correlate with WHC, SMC, and SOC at the reference site.

Parameters of Soil and Heavy Metals in Contaminated Sites
In contaminated sites (15 sites located 75°38′00.4″-75°38′29.8″N and 13°43′20.0″-13° 49′0.3″E; altitude: range 624-637 m msl), the soil temperature ranged between 25 and 35 °C. Circumneutral pH ranged between 5.6 and 7. Table 4 represents the statistics of the parameters of soil and heavy metals. The SMC and SOC content remained low, while the WHC ranged between 21 and 49%. The heavy metal concentrations in contaminated sites showed a considerable low variance, and the data were normally distributed. Except the concentrations of Cu and Cd, the values of all other parameters recorded in this study were found within the range as detailed for typical uncontaminated soils (Nagajyoti et al., 2010). The Cu and Cd concentrations were found to be high. The Ni content showed a strong positive correlation with Cu, Zn, Fe, and Cr. An increase in Ni content increased the concentration of these metals. An increase in Cr concentration showed an increase in Zn and Fe concentrations. An increase in Fe concentration showed an increase in Zn and Mn concentrations, but Zn showed a significant positive correlation with Cu. The correlation analysis between the concentration of heavy metals and the quantity of synthetic fertilizers or pesticides applied did not yield significant results except for the association between synthetic fertilizers and cadmium (r = 0.594, p = 0.019). Furthermore, the pH of the soil did not show any correlation with other parameters of soil or heavy metal concentration.

Heavy Metal Concentration in Contaminated Sites
All recorded heavy metals showed a clear increase in their concentrations at the contaminated sites (Fig. 2). The magnitude of increase differed among different heavy metals. Table 6 represents the results of ANOVA and the parameters of soil and heavy metals between the reference and contaminated sites. The WHC, SMC, Zn, Fe, and Ni showed statistically significant differences between the sites. On an average, the SMC and WHC of contaminated rice paddy soil showed 4.7-and 1.5-fold increase in their concentrations compared to the reference site and were found to be statistically significant (p < 0.05). However, the SOC content slightly decreased at the contaminated sites (0.9% of the reference site), and the differences from the reference site were insignificant. All heavy metals showed clear increase in their concentrations at contaminated sites, representing 1. respectively, the differences were marginally significant (p < 1). This could be due to the high variance in the data. Mn, Cr, Pb, and Cd, though represented at higher concentrations, did not show statistical significance. Among these, the values of Mn, Cr, and Pb also had high variance in their data. Therefore, it is apparent that Cu, Zn, Mn, Fe, Cr, Ni, and Pb showed a strong build-up in their concentrations in contaminated sites. The correlation analysis revealed a maximum significant association of heavy metals (Zn, Fe, Cr, Ni, and Cd) with SMC. This was followed by the SOC of the soil (with Fe, Cr, Ni, and Cd) and the WHC of the soil (with Cu, Pb, and Cd). PCA was performed to determine the soil parameters that has dominant influence on the heavy metal concentrations of soil. Table 7 details the components of the PCA and eigenvalues. The test revealed a significant (p = 0.001) value for the Bartlett test of sphericity for soil parameters with a KMO = 0.484 and χ 2 = 128.946, df = 55, while data on the parameters of the soil and heavy metals with the quantity of synthetic fertilizers or pesticides did not yield significant results (KMO sampling adequacy = 0.317). Thus, data on the synthetic fertilizers or pesticides were not considered to process PCA.
It was found that three principal factors with eigenvalues greater than 1, Kaiser rule, accounts for 77.944% of variance in the data. The first principal component (PC1) explained 40.259%

Discussion
In the present study, the reference site has a low concentration of heavy metals (except Cu) comparable to those of heavy metals reported for agricultural soils (Nagajyoti et al., 2010). At the reference site, SMC, SOC, and WHC were low, and these parameters did not show an association within themselves by correlation. The pH of the soil is considered the most important factor regulating the solubility and mobility of heavy metals (Reddy et al., 2013;Salem et al., 2020;Škrbić & Miljević, 2002). Furthermore, it has been observed that higher pH results in lower concentrations of heavy metals in rice paddy fields (Zeng et al., 2011). In contrast, there are reports reveal that a low soil pH results in reduction in absorption of heavy metals and their corresponding increase in the mobile fraction in the soil (Škrbić & Miljević, 2002). This represents the importance of pH in regulating heavy metal concentrations in rice paddy fields. However, in the present study, both in contaminated and reference sites, the pH of the soil remained acidic to circumneutral and did not show any correlation with heavy metals. At the contaminated sites, the concentrations of heavy metals clearly increased compared to the reference site by an order of 1.2, 1.3, 2.3, 2.2, 1.8, 2.8, 1.8, and 8.5 times for Cu, Zn, Mn, Fe, Cr, Ni, Pb, and Cd, respectively (Fig. 2). An increase in heavy metal content is generally attributed to the application of fertilizer (Huang et al., 2021;Liu et al., 2007), agrochemicals (Zhao et al., 2010), natural sources (Song et al., 2018), industrial sources, and various soil amendments . In the present study, there is clear evidence of the application of agrochemicals and pesticides in contaminated sites; however, the pesticides and concentrations of synthetic fertilizers did not show a clear correlation. Furthermore, there were no other sources of heavy metals, such as industrial effluent, entering the rice paddy fields. Farmers apply natural organic materials such as organic vermicompost, humic substances, and plant extracts in addition to synthetic fertilizers. In the present study, it was observed that there was a strong correlation between only SMC-Zn, SMC-Fe, SMC-Cr, SMC-Ni, SMC-Cd, and SOC-Fe, SOC-Cr, SOC-Ni, SOC-Cd; similarly, WHC was correlated with Cu, Pb, Ni, and Cd. Based on the correlation pattern observed among the parameters of soil and heavy metals, it can be inferred that SMC, Fe, Mn, Cr, Ni, and Cd are positively loaded with PC1, i.e., with increase of SMC, also increase the levels of Fe, Mn, Cr, Ni, and Cd in investigated soil samples and vice versa; WHC, Cu, Zn, and Cd are loaded with PC2-namely, the identified level of Cu are negativity relate with WHC, Zn, and Cd, indicating that low level of WHC represents the increase of Cu in the soil samples, while WHC, SOC, Zn and Pb are positively loaded with PC3. Therefore, correlation between parameters has clearly reflected the loading pattern in PCA. A similar trend of grouping pattern between different techniques has been observed by Hégberger and Škrbić (2012). In general, heavy metals in soil are in the form of cations and are retained on soil particles by forming chemical bonds with organic or inorganic ligand ions (Zeng et al., 2017). The soil organic content is known to significantly affect the speciation and total concentration of heavy metals due to its strong complexation ability (Zhang et al., 2020). Similarly, WHC and SMC contribute significantly heavy metal transformation (Zhang et al., 2020). Therefore, in the present study, a strong correlation between the parameters of the soil and all heavy metals is observed.
Among the contaminated sites, the probable source of additional heavy metals is the application of synthetic fertilizers. However, high contents of WHC, SMC, and SOC might have retained these heavy metals at higher concentrations over the period. It has been noted that organic matter contains hydroxyl and carboxyl functional groups, which form an organic metal complex resulting in increasing metal concentrations . Therefore, the soil organic matter is identified as the principal factor, besides soil pH, that governs the HM retention capacity of soils (Lair et al., 2007). Metal ions in the soil form stable complexes with the functional groups such as -COOH and -OH present in the organic matter, and resist desorption (Guo et al., 2006). By regulating the nutrient solubility in soil solution, organic matter influences the mobility and availability of metals in the soil (Zeng et al., 2011).
In the present study, the build-up of heavy metals could be due to the influence of SOC, WHC, and SMC on the heavy metals that were induced in rice paddy soils by the application of synthetic agrochemicals. Furthermore, in multi-metal contaminated soils, it has been recorded that the presence of one metal can alleviate the immobilization efficiency of the others (Kumpiene et al., 2008). Thus, in many cases, the interacting concentrations among Pb, Cd, and Zn were recorded (Rodríguez-Vila et al., 2014). In our study, we found an increase in the Pb content with an increase in the Cd content of the soil. In these rice paddy fields, farmers used bio-organic fertilizer and biosolids as initial nutrients at the time of tilling of the soil for paddy cultivation. These bio-organic fertilizers are known to accelerate soil organic matter and enhance fertility and water retention (Wijesekara et al., 2017;Wu, 2017). Therefore, agricultural practices and types of soil nutrient supplements may influence the build-up of heavy metals in rice paddy soils by enhancing organic content and water holding capacity.

Conclusions
Our study showed that the heavy metal concentrations in cultivated rice paddy soil exceeded the concentrations of the same metals recorded for the reference site. The heavy metals clearly showed a build-up in the order of 1.2, 1.3, 2.3, 2.2, 1.8, 2.8, 1.8, and 8.5 times for Cu, Zn, Mn, Fe, Cr, Ni, Pb, and Cd, respectively, in contaminated sites. In addition to correlation among heavy metals, a strong correlation was observed between SMC and Zn, Fe, Cr, Ni, and Cd, followed by SOC and Fe, Cr, Ni, and Cd, similarly, WHC with Cu, Pb, and Cd. The evidence on the influence of WHC, SMC, and SOC on heavy metal accumulation provided in this study could be used as basic information for the heavy metal management in rice paddy cultivation.