Characteristics of the experiment site

This study was carried out at the locality El Copal, located 10 km from Irapuato, in the south-central part of the state of Guanajuato (20° 44’ 42.10” N, 101° 19´39.63” W and average altitude of 1800 m). In this region, there are extensive intermontane valleys, filled with modern continental sediments.

The climate is subhumid warm, with annual precipitation between 550 and 700 mm and average annual temperature of 18° C. The average annual evaporation is 1300 mm, with a water deficit of 600 mm, so the depth of extraction of well water dedicated to irrigation increases every year. Due to rocky bodies and their origins, irrigation water can be a source of heavy metals.

The soils cultivated at El Copal area are vertisols (^{USDA, 2014}), of dark color, formed in temperate climates under a rustic humidity regime and characterized by a high content of clay minerals (^{Eash, et al., 2008}).

Soil used

We used 1.82 m³ of soil in the layer of 0.3 m deep, of an agricultural plot with bulk density of 1.11 g cm-3, 55.96 % clay, pH of 8.17, and 2.76 % of organic matter. Irrigation at El Copal is with deep well water and its annual average concentration is 330 μg L^-1.

Experimental design

The effect of P was evaluated with a multifactorial experimental design, with a variable level of type 3x3x2. In this stage, we evaluated the concentration of As and P in the irrigation water and growth of barley over time. The levels of P in the irrigation water (PAR) were 120, 210 and 300 μg L^-1 and those of As (AsAR) were 50, 200 and 400 μg L^-1; and the development time (t) was 45 and 90 d after sowing. A control or target (B) was distilled water (As and P free) for irrigation (Table 1). The response variable was the concentration of As in the plant (AsPP). All treatments were performed in triplicate.

Plant cultivation

We conducted sowing with certified seed (var. Esmeralda) on December 20, 2009, in 57 piles of soil of 0.126 m² land, with 38 kg each. In each pile, 45 seeds were manually sown, by the staggered method (plants in parallel rows in which each plant of a row faces a hole, between two plants of the next row). Ammonium sulfate at 19 % per ha (75 kg N) was applied at the beginning and 45 d after sowing. The plants were kept in the field near the laboratory.

Irrigation

Irrigation was performed with distilled water mixed with As (Na₂HAsO₄·7H₂O 99.98% pure; Sigma-Aldrich, Mexico) and P (NaH₂PO₄ 99.00 % pure; Sigma Aldrich, Mexico) in defined concentrations for each treatment (Table 1). Na₂HAsO₄7H₂O in water dissociates into arsenate (AsO₄ ^-3), the form commonly present in irrigation water and in the soil under aerobic conditions (^{Zhao et al., 2008}). NaH₂PO₄ in water dissociates into phosphate (PO₄ ^-3), as it occurs with phosphate fertilizers in soil under aerobic conditions. Irrigation was applied at the beginning of sowing (day 0) and 30 and 51 d after sowing (dds). In the first irrigation, we applied a 40 mm thick sheet of aqueous solution (5.04 L), in the second and third irrigation we used 60 mm (7.56 L). According to the area of the piles (0.126 m²), a total volume of 20.16 L of solution with As and P concentrations was applied according to the treatments used (Table 1).

Determination of total arsenic

The AsPP was determined in each pile at 45 and 90 dds. From each pile we extracted 15 plants (with root), washed them with distilled water and dehydrated them at 70 ° C for 72 h, crushed in a mortar and passed through a 1 mm sieve. The soil As content was determined at the beginning and end of the study and in the nitrogen fertilizer.

For the digestion of the plant, soil and fertilizer samples, we used the EPA3051A/2007 method. The total As was quantified with the method EPA 6010C 2007/ICP: 0.5 g of homogenized sample were weighed, 10 mL of concentrated HNO₃ (metal trace degree, 69 %, Fisher Scientific) were added and placed in a microwave oven (CEM, Model MARSX 230/60). The digestion took place in two stages: 1) with pressure of 0.827 MPa (120 psi) at 90 °C and constant temperature for 15 min; 2) the mixture was heated at 130 °C (reached in about 5 min) and kept for 25 min. The digested samples were cooled to room temperature, filtered (Watman paper 41) and 50 mL of tridstilled water were added. The P and As content was determined in an induction-coupled plasma atomic emission spectroscope (ICP-AES, model Iris Intrepid Thermo Elemental, Thermo Scientific Company, Waltham, MA, USA), which was calibrated with one white target and five standards.

Arsenic in soil

The average concentration of As in the soil at the beginning of the study was 1 372 ± 20 μg kg^-1. This concentration was low compared to that reported by ^{Dahal et al. (2008)} (6 100 and 16 700 μg kg^-1) in surface soil of the Nawalparasi district, Nepal, and by ^{Roychowdhury et al. (2005)} (up to 19 400 μg kg^-1) in India. In the area of mining activity in Zimapán, Hidalgo, the soil As concentration was 52 000 μg kg^-1 (^{Ongley et al., 2007}) and the As accumulation in the plant tissue was 8 440 to 12 390 μg kg^-1, in some of the 62 crops, including epazote (Chenopodium ambrosoides), chayote (Sechium edule), chilacayote (Cucurbita ficifolia) and parsley (Petroselinum crispum), (^{Prieto et al., 2005}).

In our experiment the concentration of As in the soil (1 372 ± 20 μg kg^-1) was lower than in the studies mentioned, mainly because the cultivation of rice, vegetables and bulbs requires more irrigation than grains, such as barley, wheat, corn and sorghum. In this regard, ^{Mukherjee et al. (2017)} noted that the volume of irrigation water for rice production in India is 12,000 to 14,000 m³ ha^-1 and, according to ^{Perez et al. (2011)}, barley in Guanajuato utilizes a maximum irrigation volume of 7,540 m³ ha^-1. Although in Guanajuato there is a mining zone, our study area is more than 45 km away, so it is not an As source for agricultural soils; in addition, the fertilizer used did not provide As.

Arsenic in plants

The average concentration of AsPP at 90 dds was higher than at 45 dds, but in the control (B) increased 1.1 % (Figure 1). This indicates that the rate of As absorption by the plant increased and accumulated 4.7 times more As in the last 45 d (between tillering and stem elongation). Malhi et al. (2012) indicated that barley absorbs P in sigmoid form, from its emergence until the stage of grain filling. The competition of P and As explains the increase in the speed of absorption of As by the plant. The concentration of AsPP in B did not increase between 45 and 90 dds, and the plant grew keeping the As absorption rate of the soil per kg of tissue.

† [As] p - Concentration of arsenic in the barley plant.

Figure 1 As concentration in the barley plant (B: control; n = 3). 

In addition, the average concentration of AsPP in the treatments with 400 μg L^-1 of AsAR, at 90 dds, was 2.4 times higher than that of the treatments with 50 μg L^-1 of AsAR. The average of As accumulated in the plants of the treatments with 300 μg L^-1 of PAR was 6 871 ± 1051 μg kg^-1, and the average of AsPP for the treatments with 120 μg L^-1 of PAR was 11 733 ± 4 229 μg kg^-1. This represented 41.4 % less As accumulated in barley due to a higher PAR concentration. Besides, the AsPP of barley with the AS3P1 treatment was 2.2 times higher than with As1P1. Also, the As1P3 treatment showed less AsPP accumulation than As1P1 and As1P2; similarly, in As2P3 with respect to As2P1 and As2P2, and in As3P3 compared to As3P1 and As3P2. The highest accumulation of As in the plant was found at 90 dds in the treatments As3P2 (15 354 ± 91 μg kg^-1) and As3P1 (14 926 ± 67 μg kg^-1) that used irrigation water with the highest amount of As and the lowest of P. The two minor AsPP concentrations (at 90 dds) were observed in the treatments As1P3 (4 920 ± 39 μg kg^-1) and As2P3 (3 670 ± 35 μg kg^-1), which had the highest concentration of PAR. This behavior is similar to that described by ^{Saldaña-Robles et al. (2012)}.

At the beginning of the study, the soil contained 1 372 ± 20 μg As kg^-1 (As0, Figure 2) and by the end (CASFIN) 1 386 ± 2, 1 469 ± 12 and 1 568 ± 5 μg kg^-1 in the treatments with irrigation water having 50, 200 and 400 μg As kg^-1, respectively, which were increments of 1.1, 7.1 and 14.2 % in CASFIN, with a significant linear correlation (R² = 0.999) between percentage and concentration of AsAR. The piles of each treatment (same AsAR) had a low variation, with a maximum of 20 μg kg^-1. Therefore, the accumulated amount of As in the soil depended only on the AsAR concentration and not on PAR concentration. These results coincided with those by ^{Dahal et al. (2008)}, who monitored the influence of As contamination by irrigation water on alkaline soils and the intake of As by the plants in the field, and concluded that the concentration of As in the soil and plant is influenced by the content of As in the irrigation water.

† [As] s: Concentration of As in the soil at the end of the experiment.

¶ As0: Concentration of As in the soil at the beginning of the experiment.

Figure 2 Concentration of As in the soil (± SD) at the end of the barley crop (B: control). 

Plants with the same AsAR concentration showed differences in the As absorbed and different PAR concentration, suggesting that PAR affected the concentration of AsPP, but not in CASFIN. The effect of PAR is explained by its wide interval in the mass difference of As accumulated between the soil and plant (mainly at early ages), and the increased occurrence of ion exchange processes, in which As in the irrigation water greatly accumulates in terms of total mass in the soil, compared to the amount of As absorbed by the plant. However, the concentration of As in the barley plant can be up to 12 times greater than that of the soil.

The concentration of As and P in the irrigation water and as a result of the effect of t on the development of plants was significant (p ≤ 0.05) in the concentration of As in the plant. The interactions AsAR-AsAR, AsAR-t and PAR-t were also significant (Table 2).

The simple statistical effects showed that the combination of the highest concentration of AsAR (400 μg L-1), the lowest of PAR (120 μg L-1) and the highest t (90 d) tended to accumulate more AsPP (> 7 400 μg) kg-1). On the contrary, the combination of the lowest concentration of AsAR (50 μg L-1), the highest concentration of PAR (300 μg L-1) and the lowest t (45 d) showed that AsPP decreased (<3 400 μg kg-1). (Figure 3). The effect of PAR on accumulated AsPP could be due to the similarity of the chemical form of phosphates and arsenates, which would generate competition in their absorption by the plant, and in adsorption sites in the soil. ^{Lee et al. (2016)} confirmed this mechanism by describing arsenate as analog of phosphate and the access route to the rice plant, with the same transporters. ^{Smith et al. (2002)} indicated that the addition of phosphates to the soil can release arsenates and favor their absorption by the plant. In our study, we observed an opposite effect, which strengthens previous findings that show that arsenate competes with phosphate as a substrate in the nutrient intake of the plant, with greater affinity for phosphates (^{Zhao et al., 2008}; -Jiménez et al., 2012). Then, the addition of phosphate competed with arsenate for the soil adsorption sites, which in turn affected As availability for the plant that preferably incorporated P, and the soil was the medium that accumulated most of the As. ^{Bolan et al. (2013)} observed that P can inhibit the absorption of As by the plant; therefore, the presence of P in the irrigation water decreased the accumulation of As in the rice plant, according to ^{Lu et al. (2010)}.

Figure 3 Main effects (AsAR, PAR and t) of As concentration on the barley plant. 

The ASAR-PAR interaction was not significant (p>0.05) for the accumulation of As in the plant, whereas the AsAR-AsAR interaction actually was (Table 2 and Figure 4). This indicates an exponential increase of AsPP concentration dependent on the increase of AsAR concentration. The significance of the AsAR and t (p ≤ 0.05) interaction was observed in the increase of As in the plant, which on average rose 4.7 times between sowing, at 45 d and 90 d of crop development; PAR-t had a significant effect on the accumulation of As in the plant because at 45 d of development, AsPP accumulated in the plant almost in the same amount as any PAR concentration, and at 90 d, the higher the concentration of PAR the lower that of AsPP (Figure 4).

Figure 4 Interaction effects of AsAR-PAR, AsAR-t and PAR-t.