Cadmium and phosphorus accumulates in soil under intensive cultivation in Turkey

Keywords:

• cadmium
• phosphorus fertilizer
• intensive and dry agricultural sitws
• soil

Introduction

Continuous population growth and increasing standard of living in the world increase the use of mineral fertilizers in agricultural areas. Since cultivated areas are not expected to expand, it will be necessary to intensify agricultural production, wherein fertilizer will be play a vital role (Tisdale et al., 1993). Phosphorus fertilizers applied for increasing yield may cause increasing Cd levels on agricultural land. The rate of this increase depends on application rates of phosphorus fertilizers and soil characteristics (Hashemi, 1996; Taylor, 1997). Cadmium in phosphorus fertilizers derives principally from the phosphate rock which is used for P fertilizer production (Camelo et al., 1997). As phosphorus fertilizers contain higher concentrations of Cd than those present in the soil, their application, over time, increases soil Cd concentrations (Williams & David, 1973; Taylor, 1997; Laegreid et al., 1999). Conversely, Jeng & Singh (1995) reported that long-term fertilizer P application did not elevate Cd accumulation in soils. This might be the result of using P fertilizers containing lower rates of Cd, or plant uptake and removal of Cd by leaching (Mann et al., 2002). Cadmium is a heavy metal that can be taken up from soils by plant roots, and increasing intake of cadmium with agricultural foods may affect human health. Therefore, there is a need to limit Cd levels in food (Loganathan & Hedley, 1997). Cadmium can be bound in soil by simple electrostatic forces, or by association with metal oxides, carbonates, and organic matter (Martin et al., 2002). Cadmium solubility in soil depends on soil pH and other factors. Soil Cd concentration decreases with increasing soil pH (Christensen, 1989). The high pH and CaCO₃ contents in soils may elevate adsorption rates for P and Cd by the formation of a surface complex of CaCO₃₋P and Cd (Cole & Olsen, 1959; Afif et al., 1993; Hooda & Alloway, 1998; Ramachandran & D'Souza, 1998). Long-term application of P fertilizers in excess of plant needs may raise the level of surplus P in soils. The increase in soil P content may cause increasing P loss by leaching (Hooda et al., 1999). Determination of polluting factors in agricultural areas and taking necessary precautions against those factors are vital for the future of sustainable agriculture.

There are many researches related to soil available and total P contents in agricultural areas in Turkey. However, little research has been carried out in relation to Cd content in the same areas. Therefore, the objective of this study is to investigate the effect of long-term application of P fertilizers on soil Cd and P levels on an intensive agricultural site (IAS) and a dry agricultural site (DAS).

Material and methods

Site description

The research area is located between Tokat (40°19′ N, 36°28′ E) and Turhal (40°17′ N, 36°068′ E) in middle Black Sea region of Turkey where the mean elevation is 580 m. (Fig. 1). The average annual precipitation is 436 mm and mean annual air temperature 13°C in the region (State Water Works 1974).

Fig. 1. Map of sampling sites from Kazova Plain in Tokat, Turkey.

This study was conducted on an intensive agricultural site (IAS) and a dry agricultural site (DAS). The two sites are adjacent to each other, and IAS has 0–2% slopes while DAS has 2–6% slopes. IAS covers approximately 22,000 ha and DAS 8,000 ha (Durak, 1989).

Irrigation channels (both right and left site of the plain) were installed in 1960–1965. Until 1975, open ditches drained the plain; in 1975 tile drains were also established in the Kazova plain, Tokat. While IAS has been under irrigation since 1965, DAS has been non-irrigated production for more than 60 years due to greater slope and insufficient water resource in the area.

The soils were classified as ustifluvent at IAS, and ustocrept at DAS (USDA soil taxonomy) by Durak (1989). While vegetables, especially tomatoes and beans, sugar beet and grain are grown at IAS where regularly cultivated and irrigated, mono agricultural (wheat) production is more common at DAS. Generally, suggested fertilizer-P rates in the research region varied from 30 to 150 kg P ha⁻¹ yr⁻¹ at IAS, and 26 to 35 kg P ha⁻¹ yr⁻¹ at DAS, depending on crop species and soil characteristics (Genç, 1985; Ulgen & Yurtsever, 1995). These fertilizer-P rates have been applied for more than 60 years in IAS and DAS. However, fertilizer-P application rates in practice often exceed the recommended rates in the region, varying from 39 to 200 kg P ha⁻¹ yr⁻¹ at IAS, depending on crop species. Cultivation of indeterminate tomato types has been continuously increasing at IAS since 1994. Phosphorus fertilizers as TSP (triple superphosphate) and DAP (diamomonium phosphate) have been intensively used at the experimental sites for about 60 years. TSP₁, TSP_2, DAP₁ and DAP₂ fertilizers are produced in different factories and contain 5.6–15.2 and 24.9–27.2 mg Cd kg⁻¹, respectively; other fertilizers such as compound fertilizers, calcium ammonium nitrate and urea contain lower levels of Cd (Saltali et al., 2003, unpublished data).

Soil sampling and analyses

Soil samples were randomly taken from 0–25 cm soil depth at IAS (n=18) and DAS (n=12) at the Kazova plain.

Soil samples were air-dried, sieved using a 2 mm mesh size stainless steel sieve and stored in polyethylene bags until analysed. All the plastic and glassware were cleaned by soaking in dilute HCl (1+9) and were rinsed with deionised water prior to use. Particle size distributions of the soil samples were determined by using the hydrometer method (Bouyoucos, 1951). Soil pH and electrical conductivity (EC) were determined in a 1:2.5 soil: water (w:v) extract (Richards, 1954), and organic matter (OM) content using the Walkley–Black method (Nelson & Sommers, 1982). Carbonates were measured by the calcimeter method (Nelson, 1982). Cation exchange capacity (CEC) was determined by saturating the soil samples with sodium acetate (Rhoades, 1982).

Total P and available P were measured by the (vanado)molybdate-digestion method and the Olsen method, respectively (Olsen & Sommers, 1982), using a UV spectrophotometer. The soil samples were extracted with diethylenetriaminepentaacetic acid extracting (DTPA) solution adjusted to pH 7.30 with hydrochloric acid (HCl), and available Cd in extracts was analysed by graphite furnace AAS (atomic absorption spectrophotometry). To determine total soil Cd, the samples were extracted by wet digestion using HNO₃, HCOI₄, HF and H₂SO₄ acids (suprapure quality, Merck) for total 31 min, and the digests were diluted with double deionised water and filtered through Whatman No. 42 filter paper. The Cd in filtrates was analysed by flame AAS (Baker & Amacher, 1982). A Perkin Elmer Analyst 700 model AAS with deuterium background corrector and graphite furnace was used in this study. The operating parameters for Cd were set as recommended by the manufacturer. Standard calibration solutions and some samples were measured to control continued accuracy of Cd analyses at the beginning, during and at the end. Each value was evaluated as the mean of three determinations.

Some physical and chemical properties of soils at IAS and DAS were compared with t-test, and correlations between total soil P and Cd contents, and available P and Cd concentrations at two sites were calculated with the Stat Most computer program (Stat Most, 1995)

Results and discussion

Summary statistics of some physical and chemical properties of the soils at both sites are given in Table 1. While differences between soils at the intensive agricultural site (IAS) and dry agricultural site (DAS) were significant (P<0.01) in terms of pH, EC, OM %, silt and sand % values, differences in terms of CEC and clay % values were not significant (Table 1). Greater standard error was observed soil EC values at both sites, and this was attributed to landscape and soil characteristics.

Table 1. Summary statistics of some physical and chemical properties of the intensive and dry agricultural sites

	Intensive agriculture site (n=18)					Dry agriculture site (n=12)
Soil properties	Min.	Max.	Mean	CV	SE	Min.	Max.	Mean	CV	SE	P
pH (1:2.5)	7.24	8.20	7.80	2.60	0.048	7.72	8.24	8.0	2.32	0.05	<0.01
EC (dSm^−1)	0.41	2.20	0.82	47.2	91.2	0.21	0.68	0.34	35.0	34.3	<0.01
OM (%)	1.50	2.31	1.89	13.0	12.9	1.04	1.86	1.42	21.1	0.09	<0.01
CEC (Cmol kg^−1)	15.0	39.0	24.2	23.5	1.34	12.0	30.0	24.0	20.04	1.41	ns
CaCO3 ^2− (%)	5.10	11.2	7.34	22.6	0.39	5.0	21.0	9.75	43.9	1.24	<0.05
Clay (%)	24.0	40.0	31.9	16	1.20	20.0	44.0	31.8	23.5	2.16	ns
Silt (%)	18.0	46.0	32.0	20.7	1.57	18.0	34.0	24.3	22.6	1.59	<0.01
Sand (%)	22.0	50.0	35.0	18.8	1.56	32.0	58.0	45.5	12.9	2.23	<0.01
n: Sample size; CV: Coefficient of variance; SE: Standard error; ns: Not significant (P>0.05); P: Level of significance calculated by t-test. OM: Organic matter%; EC: Electrical conductivity; CEC: Cation exchange capacity; CaCO3: Calcite%.

It was reported that intensive agricultural land use may change soil physical and chemical properties such as organic matter, pH, CEC or soil structure (Lal, 1985; Jaiyeoba, 2003).

Results of the mean total cadmium (Cd) and phosphorus (P) levels of the soil at two different sites are shown in Fig. 2 and Table 2. Total Cd contents were significantly (P<0.01) greater at IAS compared to DAS, and the mean total soil Cd content was higher at IAS relative to DAS (Table 2). These results indicated that greater fertilizer-P application was the major source of Cd input in soils in the study area (IAS). There are no factories, heavy industrial activities or heavy vehicle traffic, which can be sources of Cd pollution on arable land in the region. Total soil Cd contents were positively (P<0.05) correlated with total soil P contents at IAS. However, statistically significant relationships between soil Cd and P contents at DAS were not found.

Fig. 2. Contents of (a) total P, (b) available P at IAS and DAS: (c) total and available Cd at IAS.

Table 2. Total and available soil Cd and P (mg kg⁻¹) contents in Intensive agriculture site (IAS) and Dry agriculture site (DAS)

	Intensive agricultural site (n=18)					Dry agricultural site (n=12)
	Min.	Max.	Mean	CV	SE	Min.	Max.	Mean	CV	SE	P
Total Cd	0.22	0.52	0.32	27	0.021	0.09	0.21	0.153	28.8	0.013	<0.01
Available Cd	0.024	0.106	0.050	46	0.001	bdl	bdl	–	–	–	–
Total P	436	1350	823	26.8	52.6	444	860	664	19.0	36.5	<0.05
Available P	17	33	25.7	19.0	1.13	3	8	6.08	29.3	0.51	<0.01
n: Sample size; CV: Coefficient of variance; SE: Standard error; bdl: Below detection limit; P: Level of significance calculated by t-test.

Mann et al. (2002) observed significant positive relation (P<0.01) between total soil Cd and P levels on fertilized and unfertilized fields. After application of phosphorus fertilizers containing Cd to soils, Cd can be immobilized so that it can accumulate on the soil surface (Mulla et al., 1980; Mortvedt, 1987), and long-term P fertilizer application generally increases Cd levels in soil (Camelo et al., 1997; Taylor, 1997). To reduce Cd input and to maintain Cd levels at a minimum level, phosphorus fertilizer containing lower levels of Cd must be applied.

Total P contents ranged from 459 to 1350 mg kg⁻¹, and 444 to 910 mg P kg⁻¹ at the IAS and DAS, respectively. The total soil P contents were significantly (P<0.05) higher at IAS compared to at DAS (Table 2). There were significant (P<0.05) relationships between total soil P contents and total Cd and available P levels at IAS. This could be the result of the higher P fertilizers application at IAS relative to DAS, although the crops grown on IAS (tomatoes and sugar beet) removed the highest amounts of P from the soil.

Hooda et al. (2001) compared sites receiving high rates of P (70 kg P ha⁻¹ yr⁻¹) and fields receiving lower rates of P (26 kg P ha⁻¹ yr⁻¹). They found the greatest total P increase at a site with highest P inputs.

The Cd concentrations in the soil solution were in the range of 0.024–0.106 mg kg⁻¹, whereas available Cd concentrations at DAS were below the detection limit (P<0.001). It was seen that concentrations of available Cd were higher at IAS relative to DAS (Table 2). There was a weak linear relationship between available and total Cd (P<0.05) contents at IAS. This indicates that increasing total soil Cd content via P fertilizer application probably increased the available Cd concentrations at IAS. Similar results were reported where the elevated total soil Cd level increased the available Cd concentration, and following P fertilizers application the available Cd concentrations in soils increased still further (Taylor, 1997; Mann et al., 2002). Hashemi (1996) reported that increasing levels of P fertilizers application increased the available Cd and P concentrations in experimental soils. In this study, soil pH and CaCO₃ were significantly (P<0.01, P<0.05) higher at DAS relative to IAS (Table 1), and the higher soil pH, CaCO₃, and applications of the lower rates of P fertilizers were expected to minimize the available Cd concentrations.

The differences in the available P concentrations at both sites were significant (P<0.01), (Table 2). The mean available P concentration at IAS was almost four times higher than the mean concentration at DAS. The higher available P at IAS was probably a result of high rates of P applications. Brohi et al. (2002) pointed out that the mean available P concentration was 46 mg kg⁻¹ in soils of tomato fields in Kazova plain. Increasing P concentrations in soil was directly related to P input to the soils, and the high rates of P fertilization increased available P concentration (Hooda et al., 2001). Generally, the soils in study area had a loamy texture, and were artificially drained. Therefore, the elevating P concentration can increase the potential for P loss via leaching and surface runoff (Sharpley, 1995; Loganathan & Hedley, 1997; Sims et al., 1998; Hooda et al., 2001) and cause eutrophication (Turtola & Jaakola, 1995; Karaman et al. (2001) found that the available P deposited in the subsoil (30–60 cm) compared to the topsoil (0–30 cm), and they suggested deep ploughing every four to five years in the Kazova region.

Conclusion

As Cd input exceed output, soil Cd contents have clearly risen at IAS relative to DAS. The mean soil Cd content at IAS was 2.1 times higher than that at DAS. To decrease Cd inputs to the soils, P fertilizers containing the lower rate of Cd can be suggested for areas where cultivated crops require high amounts of phosphorus such as tomatoes and sugar beet. The average available P content was approximately four times higher at IAS relative to DAS. The increases in soil P contents, particularly in available soil P contents, can enhance risk for P loss via subsurface and surface runoff in the research region because of artificially drained soils and high rates of available P. Available P contents in soil can be maintained by balanced fertilization. Therefore, P fertilizers must be applied to the soil according to the crops’ needs for P and available soil P status for sustainable agriculture. Future studies should be carried out to investigate Cd and P downward movement and losses by leaching in the region.

Additional information

Notes on contributors

Kadir Saltali *

Saltali, K., Sari, H., Mendil, D. and Altin, S. (Department of Soil Science, Faculty of Agriculture, University of Gaziosmanpasa, TR-60240 Tokat, Turkey and Department of Chemistry, Faculty of Science and Art, University of Gaziosmanpasa, TR-60240 Tokat, Turkey). Cadmium and phosphorus accumulates in soil under intensive cultivation in Turkey.

Notes

Saltali, K., Sari, H., Mendil, D. and Altin, S. (Department of Soil Science, Faculty of Agriculture, University of Gaziosmanpasa, TR-60240 Tokat, Turkey and Department of Chemistry, Faculty of Science and Art, University of Gaziosmanpasa, TR-60240 Tokat, Turkey). Cadmium and phosphorus accumulates in soil under intensive cultivation in Turkey.