Superphosphate fertiliser application and cadmium accumulation in a pastoral soil

Figures & data

Table 1. Measured cadmium (Cd) concentration in SSP (phosphorus, P) fertiliser made of phosphatic rock samples from Nauru, Christmas Island and North Carolina.

Location of phosphatic rock	Cd concentration in rock	Cd concentration in fertiliser	Source
	(mg kg^−1 )	(mg Cd kg^−1  P)
Nauru	87	558^a	David et al. (1978)
Nauru	100	641^a	Syers et al. (1986)
Nauru	92^b	590^a	Pulsford (1989)
Nauru	88	564^a	Bramley (1990)
Nauru	95	609^a	Salmanzadeh et al. (2016b)
Nauru		592 ± 42	Mean ± 95% CI
Christmas Island	43	276^a	David et al. (1978)
Christmas Island	40^b	256^a	Pulsford (1989)
Christmas Island	38	244^a	Bramley (1990)^c
Christmas Island	38	244^a	Salmanzadeh et al. (2016b)
Christmas Island		255 ± 24	Mean ± 95% CI
North Carolina	48	378^c	Syers et al. (1986)
North Carolina	40	315^c	Pulsford (1989)
North Carolina	54	425^c	Bramley (1990)
North Carolina	39	307^c	Bech et al. (2010)
North Carolina		356 ± 89	Mean ± 95% CI

Note: Also shown are sources of the measurements.

^aP content of phosphatic rock was 15.6% (David et al. 1978; Syers et al. 1986).

^bMean estimated from the range of the results.

^cP content of phosphatic rock was 12.7% (Syers et al. 1986).

Table 2. Estimated mean cadmium (Cd) input rates for periods determined by the SSP (phosphorus, P) fertiliser blends (see Table 1 for nomenclature and Cd concentration data) based on interpretations of blend descriptions provided by the Fertiliser Association of New Zealand (Anonymous 2008).

Period	P fertiliser blend	Cd input to 17 kg P ha^−1 y^−1 plots^a	Cd input to 22 kg P ha^−1 y^−1 plots^a	Cd input to 34 kg P ha^−1 y^−1 plots^a
(Years)		g Cd ha^−1 y^−1	g Cd ha^−1 y^−1	g Cd ha^−1 y^−1
1952–1975	70% Nauru, 30% Christmas Island	8.2^b	10.6^b	16.4^b
1975–1983	50% Nauru, 50% Christmas Island	7.2^c	9.3^c	14.4^c
1983–1996	34% Nauru, 33% Christmas Island, 33% North Carolina	6.9^d	8.9^d	13.7^d
1996–2003	Unknown	5.8^e	7.6^e	11.7^e
2003–2009	Unknown	3.2^f	4.2^f	6.5^f

Notes: For 1996–2003, the Cd concentration was that of a fertiliser sample manufactured in the year 2000 (Salmanzadeh et al. 2016b). For 2003–2009, the mean Cd concentration was determined from the data reported by Abraham et al. (2016) as described in the text.

^aP content of fertiliser assumed to be 9%.

^b Cd content of fertiliser was 491 mg Cd kg⁻¹  P ( Table 1).

^c Cd content of fertiliser was 424 mg Cd kg⁻¹  P ( Table 1).

^d Cd content of fertiliser was 403 mg Cd kg⁻¹  P ( Table 1).

^e Cd content of fertiliser was 344 mg Cd kg⁻¹  P (Salmanzadeh et al. 2016b).

^f Cd content of fertiliser was 191 mg Cd kg⁻¹  P (Abraham et al. 2016).

Table 3. Measured physical characteristics of soil samples collected in the irrigation trial plots (mean ± 95% CI).

Depth at bottom of sampled layer	Sand	Silt	Clay	Bulk density	Stone mass
(m)	(%)	(%)	(%)	(kg m^−3)	(kg m^−2)
0.075	22.4 ± 1.2	48.5	29.1 ± 1.3	905 ± 40	1 ± 3
0.15	22.3 ± 0.9	46.9	30.8 ± 0.7	909 ± 253	15 ± 9
0.25	23.2 ± 1.6	46.0	30.8 ± 1.6	1270 ± 257	23 ± 14
0.50	30.9 ± 7.6	38.8	30.3 ± 2.6	691 ± 186	222 ± 116
0.75	75.8 ± 13.1	12.4	11.8 ± 6.6	426 ± 165	247 ± 88
1.00	93.3 ± 1.9	3.8	2.9 ± 0.8	484 ± 183	336 ± 94

Note: The sand, silt, clay and stone mass data came from the study reported by Kelliher et al. (2013) and the bulk density data from Kelliher et al. (2012).

Figure 1. Relationships between soil cadmium (Cd) density (product of Cd concentration and the bulk density) and depth, represented by a midpoint of the sampling interval, in the plots which received one level of irrigation and three levels of phosphorus (P) fertiliser, including 0 (◊), 17 (○) and 34 (Δ) kg P ha⁻¹ y⁻¹. Also shown in panel B are the regression lines. The regression slope and intercept were: 2.52 and 4.12 for 0 kg P ha⁻¹ y⁻¹, −5.84 and 5.67 for 17 kg P ha⁻¹ y⁻¹ and −5.35 and 6.22 for 34 kg P ha⁻¹ y⁻¹.

Figure 2. Relationships between cadmium (Cd) density in the soil and depth, represented by a midpoint of the sampling interval, measured in plots which received one level of phosphorus (P) fertiliser (22 kg P ha⁻¹ y⁻¹) and no irrigation (◊) or irrigation when soil water content reached 10% (○) and 20% (Δ). Also shown in panel B are the regression lines. The regression slope and intercept were −6.05 and 6.08, respectively.

Figure 3. Relationships between the back-transformed regression estimates of cadmium (Cd) density in the soil and depth, represented by a midpoint of the sampling interval, in plots which received one level of irrigation and no fertiliser (solid curve) or phosphorus (P) fertiliser applications at 17 kg P ha⁻¹ y⁻¹ (curve with the shortest dashes), 22 kg P ha⁻¹ y⁻¹ (curve with intermediate-length dashes) and 34 kg P ha⁻¹ y⁻¹ (curve with the longest dashes).

Table 4. Measured chemical characteristics of the soil samples including soil pH and the concentrations of oxalate extractable iron (Fe), aluminium (Al) and total carbon (C) (mean ± 95% CI).

Depth at bottom of sampled layer		Fe concentration	Al concentration	C concentration
(m)	pH	(mg Fe kg^−1)	(mg Al kg^−1)	(g C kg^−1)
0.075	5.4 ± 0.1	2129 ± 428	2017 ± 384	35.9 ± 4.1
0.15	5.4 ± 0.1	2418 ± 283	2235 ± 229	29.3 ± 4.1
0.25	5.6 ± 0.1	2443 ± 173	2247 ± 150	19.6 ± 2.8
0.50	5.7 ± 0.2	2793 ± 449	2869 ± 789	10.5 ± 2.1
0.75	5.8 ± 0.2	2345 ± 721	5416 ± 782	9.5 ± 1.3
1.00	5.9 ± 0.1	1240 ± 329	4583 ± 766	6.1 ± 0.7

Figure 4. Relationship between the estimated cumulative cadmium (Cd) input to plots and the mean cumulative Cd stock (0–0.5 m samples collected in 2009). For the data from left to right, the plots received annual phosphorus (P) applications (inputs) of 0, 17, 22 and 34 kg P ha⁻¹ y⁻¹. In addition, the estimated atmospheric deposition (input) rate was 0.2 g Cd ha⁻¹ y⁻¹ (Gray et al. 2003a). An estimated 10% of the total Cd input was lost by leaching (Gray et al. 2003b), and loss in outwash of 5% of inputs for the 17 and 34 kg P ha⁻¹ y⁻¹ fertiliser treatments and 10% for the 22 kg P ha⁻¹ y⁻¹ treatment (McDowell 2010). Also shown is the regression line.

Table 5. The difference between soil cadmium (Cd) stock to a depth of 0.5 m in plots which received annual phosphorus (P) fertiliser applications from 1952 to 2009 and the mean soil Cd stock in four control plots which were not fertilised (180 ± 77 g Cd ha⁻¹, mean ± 95% CI, n = 4 plots; for 17, 22 and 34 kg P ha⁻¹ y⁻¹, n = 4 plots each).

P fertiliser application rate	Measured Cd stock–control Cd stock	Estimated Cd input–atmospheric Cd deposition
(kg P ha^−1 y^−1)	(g Cd ha^−1)	(g Cd ha^−1)
17	283 ± 56	345 ± 22
22	450 ± 62	443 ± 28
34	698 ± 222	681 ± 44

Notes: For each P fertiliser application rate, the total Cd input to soils and the 95% CI were estimated by calculations using the data presented in Tables 1 and 2. The estimated atmospheric Cd input to soils was 0.2 g Cd ha⁻¹ y⁻¹ (Gray et al. 2003a). An estimated 10% of the fertiliser and atmospheric Cd input to soils was lost by leaching beyond a depth of 0.5 m (Gray et al. 2003b), and loss in outwash of 5% of inputs for the 17 and 34 kg P ha⁻¹ y⁻¹ fertiliser treatments and 10% for the 22 kg P ha⁻¹ y⁻¹ treatment (McDowell 2010).

Figure 5. The frequency distribution of rainfall in August (mm) for the years 1952–2009 with the area under the curve equal to unity. The left-hand vertical line was plotted at 32 mm, the mean potential evaporation for August. The right-hand vertical line was plotted at 79 mm, the rainfall used to estimate drainage from the soil in August (47 mm = 79 mm–32 mm) for Burn’s equation.

Table 6. Measured cadmium (Cd) stock in the soil as a percentage of the cumulative stock to a depth of 0.5 m in plots which received annual SSP fertiliser applications from 1952 to 2009 (mean ± 95% CI, n = 8 plots).

Depth at bottom of the sampled soil layer	Measured mean percentage of the cumulative Cd stock	Estimated percentage of the cumulative Cd stock
(m)	(%)	(%)
0.075	36 ± 4	37
0.15	60 ± 4	60
0.25	82 ± 4	79
0.50	100	95

Note: Also shown are estimates which were calculated using Burn’s equation as described in the text.

Figure 6. Relationship between measurements of cadmium (Cd) stock in soils to a depth of 0.075 m and to a depth of 0.5 m in plots which received no fertiliser (◊) or phosphorus (P) applications of 17 (○), 22 kg P ha⁻¹ y⁻¹ (▴) and 34 kg P ha⁻¹ y⁻¹ (Δ). Also shown is the regression line which had a SE of 130 g Cd ha⁻¹, 20% of the mean Cd stock to a depth of 0.5 m.

Figure 7. Relationships between time and the mean cadmium (Cd) concentration in soils to a depth of 0.075 m measured in plots which received no fertiliser (○ from McDowell (2012) and ● from Gray et al. (1999b) and this study) or phosphorus (P) applications of 17 (□ and ▪) and 34 kg P ha⁻¹ y⁻¹ (Δ and ▴). Also shown are regression lines fitted to the data for the plots which received 17 kg P ha⁻¹ y⁻¹ (solid) and 34 kg P ha⁻¹ y⁻¹ (dashed) as described in the text.

David DJ, Pinkerton A, Williams CH. 1978. Impurities in Australian phosphate fertilizers. The Journal of the Australian Institute of Agricultural Science. 44:132–135.

 Web of Science ®Google Scholar

Syers JK, Mackay AD, Brown MW, Currie LD. 1986. Chemical and physical characteristics of phosphate rock materials of varying reactivity. Journal of the Science of Food and Agriculture. 37:1057–1064. doi: 10.1002/jsfa.2740371102

 Web of Science ®Google Scholar

Pulsford J. 1989. The Australian fertiliser industry–some perspectives. Agricultural Science. 2:38–43.

 Google Scholar

Bramley RGV. 1990. Cadmium in New Zealand agriculture. New Zealand Journal of Agricultural Research. 33:505–519. doi: 10.1080/00288233.1990.10428451

 Web of Science ®Google Scholar

Salmanzadeh M, Hartland A, Stirling C, Balks MR, Schipper LA, Georger E. 2016b. Use of cadmium isotopes to distinguish sources of cadmium in New Zealand agricultural soils: preliminary results. Fertolizer and Lime Research Centre Occasional Report 29, Palmerston North: Massey University; p. 4.

 Google Scholar

Bech J, Suarez M, Reverter F, Tume P, Sánchez P, Roca N, Lansac A. 2010. Selenium and other trace element in phosphorites: a comparison between those of the Bayovar-Sechura and other provenances. Journal of Geochemical Exploration. 107:146–160. doi: 10.1016/j.gexplo.2010.04.002

 Web of Science ®Google Scholar

Anonymous. 2008. Report one: Cadmium in New Zealand agriculture. Report to Ministry of Agriculture and Forestry, 85 pages.

 Google Scholar

Abraham E, Cavanagh J, Wood P, Pearson A, Mladenov P. 2016. Cadmium in New Zealand’s agriculture and food systems. In: LD Currie, R Singh, editor. Integrated nutrient and water management for sustainable farming. Occasional Report No. 29, Palmerston North: Fertilizer and Lime Research Centre, Massey University; p. 13.

 Google Scholar

Kelliher FM, Curtin D, Condron LM. 2013. Soil carbon stocks in particle-size fractions under seasonally irrigated, grazed pasture. New Zealand Journal of Agricultural Research. 56:239–244. doi: 10.1080/00288233.2013.808673

 Web of Science ®Google Scholar

Kelliher FM, Condron LM, Cook FJ, Black A. 2012. Sixty years of seasonal irrigation affects carbon storage in soils beneath pasture grazed by sheep. Agriculture, Ecosystems and Environment. 148:29–36. doi: 10.1016/j.agee.2011.10.022

 Web of Science ®Google Scholar

Gray CW, McLaren RG, Roberts AHC. 2003a. Atmospheric accessions of heavy metals to some New Zealand pastoral soils. The Science of the Total Environment. 305:105–115. doi: 10.1016/S0048-9697(02)00404-7

 PubMed Web of Science ®Google Scholar

Gray CW, McLaren RG, Roberts AHC. 2003b. Cadmium leaching from some New Zealand pasture soils. European Journal of Soil Science. 54:159–166. doi: 10.1046/j.1365-2389.2003.00495.x

 Web of Science ®Google Scholar

McDowell RW. 2010. Is cadmium loss in surface runoff significant for soil and surface water quality: a study of flood irrigated pastures? Water Air and Soil Pollution. 209:133–142. doi: 10.1007/s11270-009-0186-6

 Web of Science ®Google Scholar

McDowell RW. 2012. The rate of accumulation of cadmium and uranium in a long-term grazed pasture: implications for soil quality. New Zealand Journal of Agricultural Research. 55:133–146. doi: 10.1080/00288233.2012.662901

 Web of Science ®Google Scholar

Gray CW, McLaren RG, Roberts AHC, Condron LM. 1999b. The effect of long-term phosphatic fertiliser applications on the amounts and forms of cadmium in soils under pasture in New Zealand. Nutrient Cycling in Agroecosystems. 54:267–277. doi: 10.1023/A:1009883010490

 Web of Science ®Google Scholar