Influencing Factors on Bioavailability and Spatial Distribution of Soil Selenium in Dry Semi-Arid Area

Abstract

The chemical transformation of selenium (Se) in the topsoil, especially when regarded as low to sufficient Se (with high bioavailability) in dry arid environments, has great importance in the alkaline soils to yield Se-enriched food regionally. The Se content in the highly alkaline soil of the northwest region of China has inordinate agriculture economic potential, and such soil distribution is likely to produce Se-enriched crops with distinct features. One such large area of Zhongwei was investigated for the distribution of soil Se and its bioavailability, and the influencing chemical factors of soil total Se (T-Se) and bioavailable Se (B-Se) in the agroecosystem. The results suggested that the T-Se in Zhongwei soils (mg/kg) ranged from 0.01 to 0.55 with a mean of 0.2 ± 0.08, which was lower than the average Se distribution of both China (0.29 mg/kg) and the world (0.40 mg/kg). However, the overall B-Se proportion (16%) in T-Se was adequately higher than in other Se-rich soils. Spatial distribution depicted that the T-Se was specified as deficient in 42.6% and sufficient in 55.5% of the studied area, while Zhongning county was prominent with a higher B-Se proportion (22%) in the T-Se of Zhongwei. The influencing factors, such as pH and organic matter (OM), showed significant association with B-Se, as suggested by Pearson’s correlation and multiple linear regression (MLR). Furthermore, the vertical distribution of T-Se and B-Se was higher in agricultural soil (AS) than in natural soil (NS) and can be justified in the context of their association with OM. Based on these results, the Se-fortified crops can be yielded by practices to improve corresponding influencing chemical factors of soil, especially in dry areas.

1. Introduction

Selenium (Se) is one of fewer trace elements whose soil concentrations diverge worldwide under natural conditions. It is an essential micronutrient for humans and animals [1,2,3], and 25 human selenoproteins are involved [4] with antioxidants, anti-viral and anti-cancer activities [5]. Selenium exists as ionic selenite (Na₂SeO₃) and selenate (Na₂SeO₄), solid-state Se, selenocysteine, and selenomethionine in the environment [6]. Due to narrow ranges of Se between adequate and toxic levels [7], moderate Se can have positive effects on biological functions, while Se with low and high levels can impose undesirable effects on human health [8,9,10,11,12].

The Se content in the soil was frequently found in an uneven distribution in the soil environment [13,14,15] due to soil controlling factors [16]. Worldwide, the Se content in soil varies from 0.01 mg/kg to 2.00 mg/kg (mean of 0.40 mg/kg) with different geological parameters [17,18], and the highest value reported to be 1200 mg/kg (seleniferous area) [7]. The Se concentration in soils significantly diverges in many regions with insufficient Se (Russia, Japan, Finland, India, China, and Germany), regions with excessive Se (western United States, India, China, Japan, Ireland, and Spain), and relatively few regions with adequate levels [19]. The Se concentration in Chinese soils is always a major concern, especially low levels [20]. The varied soil Se distribution in China is based on different geological factors [21]. Dinh et al., [22] reported that total soil Se (T-Se) in China varies from 0.005 mg/kg to 79.08 mg/kg and estimated that about 51% of the soil contains T-Se below or close to the deficient level (0.125 mg kg⁻¹), which was 72% earlier reported by Tan, [13,23]. Additionally, Central China has been identified as a low Se region belt based on its concentration and geology surveyed in the 1960s [24]. Previously, studies found low Se content distribution from north to south of China [25,26]. Se-deficient distribution to Se-marginal areas is spread along the geographic belt from the northeastern to the southwestern parts of China [22]. The descending order of topsoil Se concentrations in China of different regions are as follows: Northwest > South > Central > East > Southwest > Northeast > North China.

Geology controls the soil concentrations of Se and consequently influences Se content in the food chain [7]. Total soil Se depends on parent soil material [27] and physiochemical parameters in the soil [26]. Numerous studies have demonstrated that, in addition to the overall amount of Se present in the soil, the bioavailability of Se in the soil also depends on other soil physicochemical factors [28,29,30,31], which includes pH [32], organic matter (OM) percentage [33,34], and Fe/Al oxides content [30,35] In some regions of China, e.g., the Enshi prefecture of Hubei, there are forms of acidic soil (pH 5.58~6.49) rock material, while located in Ziyang (county of Shaanxi), the pH of the parent material shows that high levels of Se are found in acid-weak and alkaline (pH 6.36~8.27), which impacts soil Se availability [36,37,38]. Xu et al. [39] and Xiao et al. [40] found that the bioavailability of Se in soil decreases with the increase of pH in an acidic environment, and soil Se bioavailability increases with an increase in a moderate alkaline environment [41]. Meanwhile, studies revealed that highly alkaline soil with a pH ranging from 7.5–10.2 reduced the soil’s bioavailable Se [42,43]. Similarly, the soil OM content has a dual effect on the Se bioavailability in soil, which can either cause an augmentation or reduction of soil Se bioavailability [44]. Furthermore, the soil’s Se adsorption on soil oxides and minerals play an essential role in the fixation of bioavailable soil Se [45,46]. Therefore, there is a need to comprehend the importance of influential controlling factors in highly alkaline regions on soil Se bioavailability in the soil.

Dry climatic regions of the northwest have unique topography with different soil textures [47]. The soil is moderate and highly alkaline [42] in the northwest region of China, especially Zhongwei, with inordinate agricultural economic potential [48,49]. The soil in Zongwei can produce many Se-enriched products of various important crops, including wolfberry, also known as Go’ji (lycium barbarum) [50], jujube (Ziziphus) [46], and watermelon (Citrullus lanatus) [51,52]. In this study, T-Se contents in soil and different crops were investigated. In addition, physicochemical properties were analyzed to explore their impact on soil Se bioavailability in the arable soil of Zhongwei under natural field conditions. The main objectives of this study are to (1) determine the spatial distribution of soil Se content and estimate Se bioavailability in soil, (2) analyze the relationship and controlling factors between contents bioavailability and soil physicochemical properties, and (3) determine the distribution of soil Se within different depths. To the best of our knowledge, this study is one of the few studies conducted to investigate the effects of Se distribution in high-alkaline semi-arid soil of the continental monsoon climate and dry desert climate on soil Se distribution at the regional scale. It may have theoretical and practical ramifications for the careful management and effective use of Zhongwei’s land resources.

2. Materials and Methods

2.1. Study Area

This study was conducted in the southwest of Ningxia’s autonomous region and the upper reaches of the Yellow River. Zhongwei city, located in Northwest China, spreads over 130 km long from east to west and 180 km from north to south (36°06′05″–37°50′11″ N, 104°17′05″–106°10′42″ E). It has a semi-arid climate with typical characteristics of continental monsoon and desert climates [53]. Zhongwei city is connected to the southern edge of the Tenggeri Desert, which causes dust deposition in the area [26]. The topography is high in the southwest and low in the northeast. The Yellow River traverses the northern part of the city, and the Qingshui River is injected in the east. Zhongwei has scant precipitation (185 mm/year), intense evaporation, and 10–20 °C temperatures in the summer, together with strong winds in the winter (1774 mm). The data from the Zhongwei weather station for the multi-year average temperature is 9.37 °C. The terrain of Zhongwei city is from west to east and south to the north slope, within an altitude of 1100~2955 m. There are five types of landform classes, including desert, the Yellow River alluvial plain, platform mountain, secondary forest near Haiyuan county, and the basin’s large geomorphic unit is found in Zhongwei, where the soil types include: sierozem, loessal, aeolian sandy, anthropogenic-alluvial, alluvial, saline, and red clay [53].

2.2. Sample Collection and Preparation

Surface soil samples (0–20 cm) were collected (from March–May 2018 and April 2019) from arable soil or cultivated soil (AS) and uncultivated land or natural soil (NS) conditions in three different counties (Shapotou, Zhongning, and Haiyuan) from the southern to northern reaches of Zhongwei (Figure 1). A total of 3565 bulk topsoil samples were initially collected. Each bulk topsoil sample weighed 1000 g and was made up of soil from three different locations that were spaced apart by 30 m for the same condition. The total number of soil samples (371), which were root-associated AS samples (0–20 cm) from cultivated soils, considering the plant soil and the drip irrigation in the region fluctuation on Se distribution in AS, was 43 topsoil samples (composite soil samples) collected from NS adjacent sites to AS. Moreover, ten core samples of 100 cm from AS and seven core samples of 100 cm depth soil from NS were collected. The core samples were divided into four parts according to the depth ranges, namely 0–20 cm, 20–40 cm, 40–60 cm, and 60–100 cm. According to the classification and observation in AS and NS genesis, the characteristics of soil samples (e.g., soil type, soil texture, and pH) were observed to be similar. Considering the abundance of Se availability, a total of 306 crop samples were collected from vegetables grown (apple, Go’ji berries, jujube, rice, potato, miscellaneous grains, and watermelon) in different cultivated sites across Zhongwei to estimate total Se content in crops’ edible parts. The non-edible parts of crops were manually removed. The detailed storage procedure is described in SI, paragraph one.

Chemical Analysis of Soil Se and Plant Se

The soil samples were analyzed by the Key Laboratory of Functional Agriculture, Suzhou Institute for Advanced Studies University of Science and Technology of China, which has the national metrology and quality management system certification. Different reagents were used of analytical grade or better. The detailed methodology information can be found in Text 2 in SI.

2.3. Enrichment Factor of Selenium in Zhongwei Soil

Presently, the Se classification in cultivated soil is not based on any standards. The uniformity of enrichment in the soil is frequently determined by the enrichment factor (EF), which may be computed using the equation below [30].

EF = (Ci/Cref) sample/(Bi/Bref) background.

(1)

The ratio of Se and the reference element in the soil is (Ci/Cref) sample, while the ratio of Se and the reference element in the background environment is (Bi/Bref). Iron (Fe) was chosen as the reference element in the current study since it is one of the most stable elements in soil and is found in reasonably large amounts [54]. In general, enrichment in soils can be divided into no enrichment (EF < 1), slight enrichment (SE) (1 < EF < 2), moderate enrichment (ME) (2 < EF < 5), significant enrichment (SE) (5 < EF < 20), very high enrichment (VHE) (20 < EF < 40), or extremely high enrichment (EHE) (EF > 40; Sutherland, 2000).

2.4. Spatial Analysis for Soil Se and Distribution Prediction Model

The spatial interpolation classified kriging method (ArcGIS 10.5 Esri Inc. New York, NY, USA) was used to determine the prediction of the soil Se distribution in Zhongwei. The classified renderer employed a single-band raster layer, and the classified technique presents a themed raster by categorizing cell values. This thematic categorization is ideal for continuous phenomena, such as slope, distance, or suitability, if this divides the range into a few classes and assigns colors to those classes (Source ArcGIS 10.5). Prediction errors and standard errors were determined to assess the accuracy of soil Se prediction, bioavailability, and percentage.

The method of kriging is analogous to inverse distance weighted (IDW) in that, to anticipate an unmeasured place, it weighs the surrounding measured values. Both interpolators’ general formulas are created as a weighted sum of the data:

$$\hat{\mathrm{z}}\left({\mathrm{s}}_0\right)={\displaystyle \sum }_{\mathrm{i}=1}^{\mathrm{N}}{\mathsf{\lambda }}_{\mathrm{i}}\mathrm{z}\left({\mathrm{s}}_{\mathrm{i}}\right),$$

(2)

where Z(s_i) is the measured value at the ith location, λ_i is an unknown weight for the measured value at the ith location, s₀ is the prediction location, and N = the number of measured values [55].

2.5. Statistical Data Treatment

Initially, the Kolmogorov–Smirnov test [p(k-s) > 0.05] method of normality test was performed. Significant research has measured the contribution of the natural and anthropogenic environment using machine learning (ML) techniques. In ML algorithms, a training dataset is a critical input supporting the model’s learning ability. Creating a training dataset for supervised learning is typically a laborious effort. This study directly used ML to examine Pearson correlation and multiple linear regression (MLR) to measure the association of variables and the importance of the most fitted and contributory factor to help in soil-bioavailable selenium (B-Se) [56,57]. The coefficient of determination (R²) for RFR and LR was 0.70. The root-mean-squared error (RMSE) and Mean Absolute error (MAE) were minimum in the string of the regression models. In this study, before applying the multiple linear regression, we normalized the data on a scale from 0 to 100, which empowered us to conclude the reliable and valid association between the predictors and the explained variable. (Equation (3)).

$${\mathrm{BSe}}_{qt}={\beta }_0+{\beta }_{1}S{e}_{qt,1}+{\beta }_{2}p{H}_{qt,2}+{\beta }_{3}O{M}_{qt,3}+{\beta }_{4}CA{C}_{qt,4}+{\beta }_{5}c{y}_{jt,5}+{\beta }_{6}A{I}_{jt,6}+{\beta }_{7}A{A}_{qt,7}+{\beta }_{8}CE{C}_{qt,7}+u_{qt},$$

(3)

where β₀ the intercept, q is the observation unit, t is time, and $(u_{jt})$ is the error term [58], and where BSe represents B-Se, Se denotes the T-Se, CAC explains the CaCO₃, CY is clay, AA is amorphous iron, and CEC is cation exchange capacity. All statistical analyses were performed using ML, Origin, SPSS 23 (IBM SPSS Statistics), Stata17, and Excel 2016.

Quality Control

Several quality control methods were adopted during lab experiments, and standard errors were considered in data analysis. Certified stander material was used for lab analysis, and repetitive tests were performed to calculate the soil variables. Statistical errors were controlled by using various contents in data analysis. Spatial data analysis was controlled by the standard error of interpolation analysis in ArcGIS 10.5 Esri Inc.

3. Results and Discussion

3.1. Total Selenium Content and Spatial Distribution

The average T-Se distribution in cultivated topsoil (AS) ranged from 0.01 to 0.49 mg/kg with a mean of 0.20 mg/kg (Table S1), and T-Se concentration in natural soil (NS) adjacent to cultivated or farmland soil ranged from 0.03 mg/kg to 0.29 mg/kg with a mean of 0.13 mg/kg (Table S2), which is lower than a previously conducted study in Ningxia province near uncultivated soil of the Yellow River [59]. The average concentration of T-Se in AS and NS of Zhongwei is evidently lower than the national average T-Se (0.27 mg/kg) in cultivated soil of China [13] and average soil Se in the world (0.40 mg/kg) [7]. Moreover, it is higher than the average Se content of the topsoil in the summer (0.16 mg/kg) of Ningxia [60] for a similar area, which signifies that the T-Se content has increased slightly. The variation coefficient (CV) of T-Se distribution was 39.66% (weak variability), and an insignificant CV of about 40% indicated that Se distribution covers narrow ranges of dry arable soil of Zhongwei, which were less than black soil (65%) in North China [61].

Based on environmental importance in a low-Se area and risk assessment of seleniferous areas in China, four regional classifications of Se concentrations in soil were previously defined [23]. The detailed classification of the distribution of different regions of Zhongwei has a small margin of low-to-high soil T-Se according to the classification of national and regional standers, which can be found in SI (Table S4). The amount of Se in topsoil was divided into five grades: T-Se-deficient (<0.125 mg/kg), T-Se-low or marginal (0.125–0.175 mg/kg), moderate (0.175–0.40 mg/kg), rich (0.40–3.0 mg/kg), and excessive (>3 mg/kg) [23,62]. The soil Se content in the cultivated area was classified as Se deficient (<0.125 mg/kg; 15.6%), Se marginal (0.125–0.175 mg/kg; 27%), and Se moderate (0.175–0.40 mg/kg; 55.5%). Furthermore, nearly 43% of the soil was accounted as Se deficient to the margin (Low), while soil Se with adequate-to-high Se measured less than 2%. The soil Se distribution of Zhongwei conferring to Ningxia enrichment standers is listed in Table S5 (Ningxia People’s Education Publishing House, [63]. The mean T-Se distribution in different counties, Shapotou (ST), Zhongning (ZN), and Haiyuan (HY), were almost analogous, and no certain variation in T-Se was observed between the different counties (Table S4).

3.1.1. Soil Se Enrichment Factor in Cultivated Soil

Currently, there are no criteria to calculate total enrichment for the micronutrients such as Se in the soil. The enrichment factor (EF) is often used to calculate the uniformity of enrichment in soil method used by previous researchers [30,64]. Qiao et al. [54] used Fe as the reference element, one stable element in the soil crust and present in soil with high concentrations. The average EF values of Se in Zhongwei soil were 0.94, categorized as slight enrichment from external sources. Although, Se’s characteristics of supergene enrichment were found in studies conducted in Zhongning (ZN) county of Zhongwei. The accumulation of organic surface layer bioaccumulation and anthropogenic factors are the main reasons for the enrichment of Se in soil [65].

3.1.2. Soil Se Bioavailability Characteristics

The bioavailable selenium (B-Se) content of cultivated soil ranged from 0.002 mg/kg to 0.12 mg/kg, with a mean value of 0.034 mg/kg. The CV of bioavailable Se was 58.09%, and the proportion of B-Se in total Se was 15.9% (Table S1), which was consistent with studies conducted in Ningxia city by Zhang et al. [66]. The bioavailable Se in soil was discrete, and Se bioavailability in the center of Shapotou and Haiyuan County was found to be abundant (Table S5, Figure 2). The distribution of the B-Se proportion was not significantly different, irrespective of soil T-Se and B-Se distribution (Figure 2). The proportion of B-Se in the northeast of Zhongning county was considerably better and ranged from approximately 20% to 22% of total Se in the soil. Soil B-Se and T-Se contents exhibited a significant correlation, and the correlation coefficient was 0.42 (p < 0.01) (Figure 3). The average proportion of B-Se to total soil Se was nearly 16% in Zhongwei cultivated soils, which was adequately higher than in other areas of China with high T-Se in the soil. For instance, the soil bioavailable soil Se measured in Shaanxi soil was 11.1%, although the average total soil Se of Shaanxi province was much higher than Zhongwei (approximately 0.68 mg/kg) [41]. In China, the proportion of bioavailable soil Se in total soil Se accounted in Yongjia County (Zhejiang, China), Dianchi Lake (Yunnan, China), Hechi (Guangxi, China), and Ping’an (Qinghai, China), was 9.4%, 9.7%, 8.7%, and 3.0%, respectively, and their respective total soil Se contents were marginally higher than Zhongwei soil [31,39,67,68]. In addition, the soil B-Se in Zhongwei cultivated soil was much higher than in outer parts of Chinese soil, e.g., England (1.1–3.4%), Malawi (Africa, 2.9%), and Se excessive areas of California (USA, 10%) [4,69,70]. Additionally, previous findings in studies on the relationship between soil B-Se and T-Se are listed in Table 1.

The higher proportion of B-Se in soil was instigated by slightly high alkalinity (pH 8.27) in Zhongwei. Furthermore, soils in Northwest China are saline and alkaline, which influences the B-Se in soil [26,35]. Under highly permeable, alkaline soil conditions, selenite can rapidly change to selenate after the oxidation process, that is, a more mobile form in soil solution, and is readily available to the surface soil for up taken by the plant and results in improving Se bioavailability [71]. In addition, for plant uptake, selenate is easily assimilated and bioavailable for plants, especially under oxidizing, and neutral to alkaline conditions [72,73].

Topography and geomorphology chemical composition play major roles in the parent materials chemically produced by weathering or by erosion and sedimentation. Topographic basins often form hydrological sinks wherein the evaporation of saline water precipitates salts, which ultimately contribute to soil salinization, especially in dry soils [74]. The influenced topographical factors involved in soil Se enrichment decrease in elevation and altitude [39], although some studies found elevation had no impact on soil Se [30]. The entire relief of the Zhongwei area is high on the southern side and low on the northern side near the edge of the Yellow River, and the altitude generally changes between 1137 m and 2944 m (Figure S1). The total soil Se content in the Zhongwei area was homogenously distributed, depending on the other geological and environmental factors (Figure 2). In this study, total soil Se content is more likely associated with the parent material or geology of the area rather than topographical variations.

Precipitation is another major controlling factor of Se input in terrestrial soil, as atmospheric Se input via precipitation and dry deposition play contributing roles in parent bedrock composition and in controlling the distribution of Se source in soil [67,75]. Although environmental and climatic controlling factors, such as precipitation and drought index, were dominant in soil Se, previous studies have shown that soluble Se in the soil is easily leached by precipitation [76]. The level of Se in rainwater T-Se of China ranges from 0.1–0.2 μg Se L⁻¹, and Se depositions are estimated to be 160–320 μg m⁻² year⁻¹ [26]. In the current study, the annual precipitation was meager (~186 mm), and Se deposition was nearly negligible with precipitation. Nevertheless, slight bioavailability differences in the north and east Zhongwei might be associated with less precipitation (Figure 2). Studies on the effect of topography, elevation climatic variations, and environmental variations on the soil Se bioavailability in dry arable areas, such as Zhongwei, are still sparse and current studies emphasize B-Se in soil and its major controlling factors.

3.1.3. Spatial Distribution of Se in Zhongwei

Based on the principle of ordinary kriging interpolation, the spatial distribution of the Se contents in the soil of the Zhongwei area is shown in Figure 2. On the basics of best linear unbiased prediction (BLUP) by kriging interpolation, the Se distribution in Zhongwei soil was uniform in most of the arable areas. The soils of moderate Se were located in ST county near the Yellow River and southwest of HY county. The total Se distribution in ZN county is discrete and scant in marginal-to-moderate Se areas in comparison to other counties of ZW (Figure 2).

3.1.4. Selenium Characteristics in Different Crops

The present study divulges that soil T-Se content at the junction of the north to south sides of Zhongwei was equally distributed (Figure 2), and the Se content in crops was sufficient to be Se-rich (Table S3). Dinh et al. [22] demonstrated a holistic study of Se content in different crops of many regions in China. Previous studies have estimated Se ranges in different crops with sufficient and rich Se levels [77,78,79], and the results observed in this study corroborate with normal ranges of different crops with sufficient Se content. A high proportion of bioavailable Se in the soil leads to sufficient Se concentration in different crops (Table S3). Previous studies have also revealed that the plant Se content primarily depends on bioavailable soil [41,80]. Since the lower concentration of soil Se with a high percentage of bioavailability makes this study significant depending upon the controlling factors [81]. The Se contents measured in different crops were as follows: vegetables, apples, Go’ji, jujube, rice, potatoes, and miscellaneous grains were 0.005 mg/kg, 0.012 mg/kg, 0.022 mg/kg, 0.054 mg/kg, 0.061 mg/kg, 0.019 mg/kg, and 0.044 mg/kg, respectively. The CV of different crops varied (weak-to-average variation) (Table S3), which shows the data distribution of the different crops covering small-to-wide ranges depending upon the total soil Se concentration.

3.2. Impact of Soil Properties on Soil Se

Se absorption in the soil solutions may be a better indicator for defining Se dearth or excessiveness than the soil T-Se. Se undergoes soil adsorption/desorption [82] and precipitation/dissolution phenomena from the solution phase to the solid phase [22,71] and other processes among soil components (such as OM, clay minerals, etc.). These processes are generally controlled by soil acidic and basic conditions, which regulate Se bioavailability controlled by adsorption processes, while selenate is weakly adsorbed and forms outer-sphere, and selenite is bound as inner-sphere complexes [83]. Soil parent materials have primary effects on the soil Se apart from external input in the forms of fertilizers or anthropogenic sources [7,44]. Therefore, soil influencing factors do change the Se content in the soil and cause changes in bioavailability to the plants.

3.2.1. Influences of Organic Matter (OM) in Se Content

Soil OM plays an essential role in regulating Se bioavailability [29,67,84] since Se in the soil is bound to OM and accounts for up to 40–50% of the total Se in soil [85]. This study emphasizes that OM was a controlling factor in Se bioavailability, and the results are significant (

Table 2. Matrix of correlations of bioavailable Se and controlling factors in soil.

	Se	B-Se	B-Se (%)	pH	OM	CaCO3	Clay	AI	AA	CEC
Se	1.00
B-Se	0.76 **	1.00
B-Se(%)	−0.02	0.53 **	1.00
pH	−0.22 *	0.26 **	0.20 *	1.00
OM	0.47 **	0.22 *	−0.21 *	0.07	1.00
CaCO3	0.18 *	0.35 **	0.28 **	0.62 **	0.03	1.00
Clay	−0.04	−0.04	−0.06	0.08	−0.02	−0.01	1.00
AI	−0.07	−0.28 **	−0.16	−0.11	0.00	−0.09	−0.06	1.00
AA	−0.05	−0.06	−0.12	0.07	−0.09	−0.02	0.07	−0.05	1.00
CEC	−0.07	−0.03	0.05	−0.08	0.08	−0.07	−0.10	0.15	0.215 *	1.00

** p < 0.01 and * p < 0.05.

). Multiple linear regression models also indicated that OM was one of the strongest explanatory variables for the variation of bioavailable Se in soil (

Table 3. Results of multiple linear regression analyses for the dependence of bioavailable Se.

BSe	Coef.	St. Err.	t-Value	p-Value	95% Confidence	Interval	Sig
Se	0.734	0.054	13.59	0	0.627	0.841	***
pH	0.155	0.057	1.42	0.008	0.038	0.23	**
OM	0.252	0.069	3.10	0.002	0.249	0.055	***
CaCO3	0.117	0.055	2.13	0.035	0.008	0.227	**
Clay	−0.011	0.053	-0.20	0.841	−0.115	0.094
AI	−0.178	0.042	−4.24	0	−0.26	−.095	***
AA	−0.06	0.05	−1.22	0.226	−0.159	0.038
CEC	0.104	0.054	1.91	0.059	−0.004	0.211
Constant	−5.023	5.52	−0.91	0.365	−15.953	5.906
	Mean dependent var	26.836		SD dependent var	16.965
	R-squared	0.700		Number of observations	128
	F-test	34.713		Prob > F	0.000

*** p < 0.01 and ** p < 0.05.

). The current study, located in dry weather with minimal precipitation soil in the lowest zones of OM content in Northwest China [86], emphasizes a positive correlation with the significance of p < 0.05 of B-Se with organic matter explains 16% of the bioavailability of Se in the soil (Table 2). Previous studies documented a linear correlation between OM and Se bioavailability in soils [87,88,89]. Soil organic compounds have two diverse effects on Se bioavailability [90] and influence Se bioavailability in soil. The structure and chemical characteristics of OM define its effects on Se bioavailability [84]. Previously, positive relationships were reported between soil microorganisms’ biomass and the contents of soil organic matter [91,92]. The soil microorganisms also affect the mineralization process of OM to change the content and composition of soil OM, which promote the transformation of Se forms by interacting with soil OM [93]. Through the microbial response, organic Se could be impregnated into inorganic Se to improve the Se supply capacity of the soil, and SeO₃²⁻ also combines with soil OM to form a small amount of molecular organic Se [33]_. The relationship with available Se depends on the degree of decomposition of organic matter. The possible approaches for binding Se to OM include the formation of ternary complexes, creating anoxic zones, and sorption sites [29,94].

Table 1. Proportion of bioavailable Se in total Soil Se in previous studies.

T-Se (mg/kg)	B-Se (mg/kg)	B-Se (%)	pH	OM (%)	Clay (%)	Reference
2.00	-	8.00	8.11	4.00	-	[95]
-	-	<5	5.8	1.50	-	[96]
2.50	-	-	7.75	1.63	39.5	[97]
0.22	0.05	23.9	7.89	3.08	36.8	[81] [98]
0.14	0.04	27.1	8.06	3.44	35.7
0.23	0.06	25.1	7.85	3.72	33.1
0.17	0.04	22.6	7.25	4.43	41.3
0.52	0.01	1.55	5.77	1.19	36.3	[76]
0.41	0.04	10.1	6.61	3.03	42.1
0.53	0.08	15.8	8.14	0.85	39.6
0.38	0.03	8.90	4.75	2.44	-	[39]
0.22	0.04	19.8	8.34	0.64	18.8	[99]
0.13	0.01	3.60	8.14	0.85	39.6
0.36	0.01	1.94	6.82	5.14	11.1
0.51	0.01	1.55	5.77	1.95	36.3
0.08	0.01	12.9	7.75	1.63	39.5	[100]
0.68	0.08	13.0	6.70	2.90	-	[67]
-	-	8.50	-	-	-	[70]
0.85	0.13	17.0	-	-	-	[101]
0.20	0.02	9.70	-	-	-	[31]
0.57	0.07	13.0	7.20	-	-	[40]
0.73	0.06	7.00	7.20	-	-
0.37	0.12	3.00	-	-	-	[88]
0.59	0.01	3.40	5.60	-	-	[69]
-	-	2.90		-	-	[4]
0.25	0.02	11.10	7.90	2.24	31.4	[41]
0.167 (n = 371)	0.028	16.76	8.27	0.71	-	Present study

Table 1. Proportion of bioavailable Se in total Soil Se in previous studies.

T-Se (mg/kg)	B-Se (mg/kg)	B-Se (%)	pH	OM (%)	Clay (%)	Reference
2.00	-	8.00	8.11	4.00	-	[95]
-	-	<5	5.8	1.50	-	[96]
2.50	-	-	7.75	1.63	39.5	[97]
0.22	0.05	23.9	7.89	3.08	36.8	[81] [98]
0.14	0.04	27.1	8.06	3.44	35.7
0.23	0.06	25.1	7.85	3.72	33.1
0.17	0.04	22.6	7.25	4.43	41.3
0.52	0.01	1.55	5.77	1.19	36.3	[76]
0.41	0.04	10.1	6.61	3.03	42.1
0.53	0.08	15.8	8.14	0.85	39.6
0.38	0.03	8.90	4.75	2.44	-	[39]
0.22	0.04	19.8	8.34	0.64	18.8	[99]
0.13	0.01	3.60	8.14	0.85	39.6
0.36	0.01	1.94	6.82	5.14	11.1
0.51	0.01	1.55	5.77	1.95	36.3
0.08	0.01	12.9	7.75	1.63	39.5	[100]
0.68	0.08	13.0	6.70	2.90	-	[67]
-	-	8.50	-	-	-	[70]
0.85	0.13	17.0	-	-	-	[101]
0.20	0.02	9.70	-	-	-	[31]
0.57	0.07	13.0	7.20	-	-	[40]
0.73	0.06	7.00	7.20	-	-
0.37	0.12	3.00	-	-	-	[88]
0.59	0.01	3.40	5.60	-	-	[69]
-	-	2.90		-	-	[4]
0.25	0.02	11.10	7.90	2.24	31.4	[41]
0.167 (n = 371)	0.028	16.76	8.27	0.71	-	Present study

Furthermore, high molecular weight organic acids (HMWOAs) can form complex compounds on the surface in the high alkaline situation, which ultimately reduce Se mobility on the soil solution and causes a reduction in bioavailability [33,102,103]. The OM oxygen-containing functional group can easily chelate Se available in soil surface solution. Moreover, the OM ternary complexes may increase Se mobility [90]. Wang et al. [24] revealed in their study that the northwest of China has Se-adequate areas as arid and semi-arid areas in China and is considered to have the least precipitation and the sturdiest evaporation. These soils in the region are under strong oxidizing and alkaline conditions. The soil Se has accumulated in the basin and mountainous slopes as a result of hydrological and aerial transportation. In these developments, Se is found in alluvial plains in the form of selenate in the floodplain and the inland basin. The soluble selenate reacts with organic matter and is retained in the soil. In contrast, many studies have specified a positive correlation between OM with B-Se [29,40].

Additionally, amorphous iron (AI) is the most active iron/aluminum oxide [41], and Se can form stable inner complexes by coprecipitating with iron hydroxides [18], which ultimately absorb SeO₄²⁻ and reduces the bioavailability of soil Se. The current study emphasizes the same mechanism in soil Se availability (Table 2). Furthermore, studies revealed that the adsorption capacity of iron/aluminum oxides reduces in highly alkaline soils, and with acidic conditions (pH ≤ 6.0), it inverses the process of Se adsorption in soil oxides [39]. Therefore, by controlling the alkalinity in Zhongwei soils with agronomic practices such as tillage, the organic material amendment can produce Se-enriched soil and Se-enriched crops.

3.2.2. Influence of pH on Se Content in the Soil

Selenium oxyanions have more retention in acidic soils in comparison with alkaline calcareous soils [104]. Most alkaline soils adsorb the least amount of Se, most likely due to increased Se solubility [6,105]. Soil pH in Zhongwei is typically alkaline to highly alkaline (mean 8.27), and under well-aerated conditions, inorganic Se is mainly found in the form of selenate (SeO₄²⁻), which is hard to adsorb on the soil’s solid surface and is not only readily available to plants but vulnerable to lose by leaching [67]. The result of this study is consistent with the loss of Se with high alkalinity of the soil, which limits the bioavailability of soil (Table 3), increases the concentration of selenate on the surface, and has a good impact on the percentage of B-Se (Table 2). Furthermore, in an alkaline soil environment, the methylation of soil Se within a definite pH range intensifies with an increasing pH, promoting volatile dimethyl compounds that increase the Se mobility and enhance losses of Se in surface soil [106,107].

3.3. Verticals Distribution of Selenium in AS and NS

The concentration of Se in cultivated soil (AS) and natural soil (NS) is accumulated within the depth of 0–40 cm, and the average T-Se content was 0.32 mg/kg and 0.19 mg/kg, respectively. It depicted the concentration of T-Se and B-Se up to a depth of 0–40 cm controlled by OM in the soil (Figure 4a,b). However, with a depth of 40 cm to 100 cm, T-Se and B-Se construction normally decreases by 0.13 (mg/kg) and 0.014 (mg/kg), respectively, which is low Se in the soil according to local standers of Zhongwei (Table S4).

Studies revealed that T-Se in topsoil mainly comes from wet and atmospheric dry deposition [108]. Therefore, the vertical distribution of the soil Se might be affected by wet deposition and ultimately controlled by physicochemical parameters. The northwest Se-adequate area is the driest region in China, with the lowest rainfall and the most intense evaporation. The soil types in the area are mostly desert and alpine. These soils have high oxidizing and alkaline characteristics.

Agronomic measures can alter soil Se bioavailability by naturally and artificially controlling pH and OM [22]. Tillage and vegetation cover can change soil Se bioavailability by regulating soil Se [40,109,110] showed that soil’s physical and chemical properties could be improved by planting secondary forests in Ningxia, a semi-arid region, for several years, and soil quality can be improved significantly. For irrigation practices, soil Se in Zhongwei can be lost in the runoff from surface soil by flood irrigation because of the high alkalinity in the soil. Consequently, apposite irrigation practices (e.g., drip irrigation, sprinkler irrigation, etc.) can reduce the percolation of soil Se [48,66,111]. Effective fertilization practices of organic and inorganic fertilizers (i.e., N, P, and S) can substantially improve soil Se bioavailability [112,113,114,115]. The application of fertilizers can regulate the bioavailability of Se. However, the mechanism involved with Se bioavailability after applying fertilizers is more complicated [22]. Agronomic biofortification can be an agriculture strategy used by food companies as a cost-effective method in Zhongwei [42] to produce high-Se functional products that contain more bioavailable Se [116]. Since soil Se levels were limited at the study site, it is feasible to reintroduce residual plant material, which contained additional plant Se.

Further studies are required to know about the soil mechanism in Zhongwei and other mineral compositions of Zhongwei soil, which controls soil Se. The composition of OM in dry soil needs to be investigated in different conditions and concentrations to understand the mechanism of B-Se percentage variations in soil.

4. Conclusions

This investigation is one of the few studies to report soil Se distribution in the dry arable area of Zhongwei and showed that B-Se was directly correlated to the soil T-Se, which is controlled by the soil’s physiochemical properties. The results of this study suggested that the spatial distribution of soil Se from north to south of Zhongwei was evenly distributed, irrespective of soil Se bioavailability, which certainly changed in some cultivated sites. The study also indicates that the high Se bioavailability proportion produced sufficient-to-moderate Se crops. Moreover, bioavailable Se was primarily influenced by physiochemical factors, such as pH, and OM, more contrarily than reported in previous studies because of high soil alkalinity. Our observation concluded that OM and pH were the main influential factors in increasing soil Se bioavailability, which can hold for crop biofortification practices grown on similar soil conditions. This study presents novel data on soil B-Se information for making appropriate soil Se input recommendations to meet the adequacy of Se bioavailability for plants in dry areas such as Zhongwei. Additional studies are needed to provide a scientific approach for regulating the B-Se in highly alkaline soil and Se organic fractions in dry regions such as Zhongwei.