Can the application of rare earth elements improve yield and decrease the uptake of cadmium in ryegrass-dominated pastures?
The rare earth elements (REEs) are a chemically similar group of 17 elements, including 15 elements with an atomic number from 57 (lanthanum; La) to 71 (lutetium; Lu) as well as scandium (Sc, 21) and yttrium (Y, 39). All the elements (except promethium; Pm) exist naturally in soil parent materials and total concentrations in soil vary from 16 to 219 mg [kg.sup.-1] (Hu et al. 2006). Rare earth elements are usually found as flurocarbonates, phosphates, silicates and oxides. Hence, commercially available rock phosphates contain REEs (Otero et al. 2005), which depending on their composition and use will influence soil REE concentrations (Tyler 2004).
Until recently, little attention had been paid to REEs as they were thought to be neither essential for life nor toxic. However, several researchers have purported to find significant increases in the yield of several crops. There have been several mechanisms given as causes such as: (i) increasing nutrient (N, phosphorus [P] and potassium [K] uptake, Liu et al. 1997); (ii) stimulating enzymes involved in N processes within the plant or in soil microbes (Liu et al. 1997; Zhu et al. 2002); (iii) increasing chlorophyll production (Guo 1993); and (iv) improving plant resilience to cold tolerance or drought (Guo et al. 1988). In addition, Wang et al. (2006) summarised several studies linking the application of REEs to decreased heavy metal uptake. However, there have also been many studies that may have been compromised by inadequate design or interpretation. For instance, Liu et al. (2006) reported yield increases of 10-24% for vegetable crops, but used control plots that were located in a different province to the treatment.
Studies that have reported better crop yields when applying REEs have a single commonality of low application rates, generally <10 kg [ha.sup.-1], and more commonly <1 2 kg [ha.sup.-1]. In contrast, studies such as those by Xu and Wang (2007) and Xie et al. (2002) have both demonstrated that rates >10 kg [ha.sup.-1] of REE application decreased yield in rice and maize. Consequently, there has been concern that long-term application of some REEs may harm soil and aquatic ecosystems (Boger et al. 1997; Wang et al. 2001; Brioschi et al. 2013), or bioaccumulate along the food chain (Liang et al. 2005). However, the lethal dose for 50% of the population ([LD.sub.50]) for La is similar to table salt (Wald 1990). Rambeck and Wehr (2005) point out that these results are likely to translate to other REEs, since similar chemical and biochemical behaviour span all REEs. Hence, while toxicity should be confirmed for each REE, their application at low rates to crops is unlikely to result in a toxicity issue.
In New Zealand, agriculture is dominated by ryegrass-white clover pastures, grazed by cattle, red deer and sheep. Farmers seek methods to produce pasture more efficiently, while also being conscious of potential environmental constraints such as the enrichment of topsoils with Cd arising from the application of P-fertilisers (Cadmium Working Group 2011; McDowell et al. 2013). We therefore hypothesised that the addition of REEs to soil at low rates (<2 kg [ha.sup.-1]) may benefit the yield of ryegrass-white clover pasture, while decreasing Cd uptake in foliage. To test this hypothesis we conducted a soil survey to determine background REE concentrations, and a series of glasshouse studies to determine environmental and soil conditions that affect ryegrass yield and Cd uptake following REE application, and can be used to explain responses tested in the field.
Materials and methods
Experimental studies were conducted in four stages: (i) a survey of New Zealand soils and fertilisers was conducted to determine the distribution and concentration of REEs in a range of New Zealand soils, and therefore the rate necessary to enrich soil REE concentrations above a natural background concentration; (ii) screening of a range of REEs, applied at a low rate, for their efficacy in improving ryegrass yield and decreasing Cd uptake in pot trials in a glasshouse; (Hi) further testing of REEs for yield and Cd uptake in the glasshouse under different soils, soil moistures and application rates and finally; (tv) testing of the best performing REE(s) from the glasshouse studies on existing ryegrass-white clover pastures in the field.
Soil and fertiliser survey
Ninety-nine soil samples were obtained from a previous trial (McDowell et al. 2013). Soils were classified by land use (dairy or sheep and beef) and their parent material (peat, sedimentary or volcanic). Rocks used in the manufacture of superphosphate were also obtained from Ballance Agri-Nutrients, Tauranga, New Zealand. Both soils and P-fertilisers were digested in concentrated nitric acid according to the method of Zhang and Shan (1997) and Cd and REEs (Ce, Dy [Dysprosium], Er [Erbium], Eu [Europium], Gd [Gadolinium], Ho [Holmium], La, Lu, Nd [Neodymium], Pr [Praseodymium], Sm [Samarium], Tb [Terbium], Th [Thorium], Tm [Thylium] and Yb [Ytterbium]) measured with an Inductively Coupled Plasma Optimal Emission Spectrophotometer (ICP-OES) with a Cetac U5000AT ultra sonic nebulizer.
Glasshouse screening of REEs
The soil chosen for the pot trial was a Warepa silt loam soil classified under the New Zealand soil classification system as a Pallie soil (USDA Classification: Udic Haplustept). Soil was obtained by scraping off the first 5 cm of topsoil (to minimise the influence of past fertiliser applications) from a permanent grassland and sampling the 5-10 cm layer. Intact soil cores were taken of this depth and bulk density determined using the method of Drewry et al. (2000). Extracted soil was air-dried for 4 weeks, crushed to <4 mm and thoroughly mixed. A subsample was crushed to <2 mm for analysis of anion exchange capacity (ASC), cation exchange capacity (CEC), Olsen P, pH, quick test K, and total C and N using the methods outlined in Blakemore et al. (1987). Mean chemical and physical characteristics of soils used in this paper are given in Table 1.
Approximately 3 kg of air-dried soil was then put into each pot and adjusted to 40% of available soil moisture (ASM; Grewal et al. 1990) determined as the difference between field capacity and wilting point: field capacity was 36% v/v and wilting point 13% v/v. Treatments were arranged in a randomised block design with a buffer-pot included between blocks and the outside of the trial.
After 2 weeks, 50 ryegrass seeds were sown per pot, allowed to germinate and culled to an average of 30 plants per pot to give both the same potential for foliage per pot (without treatment effects). Thirty days later foliage was cut 5 cm above the soil surface (to leave a residual dry matter (DM) equivalent to 1300 kg [ha.sup.-1]) and harvested. A temperature controlled greenhouse was used to grow the plants with temperature maintained at 22[degrees]C during the day and 10[degrees]C overnight. The pots were watered with an automatic overhead irrigation system that delivered ~2.5 mm twice per day. This was sufficient to keep the soil at ~30-40% ASM. During this establishment period, artificial lighting was used to extend the normal daylight period to 12 h.
An additional harvest occurred another three weeks later, after which it was assumed that the effect of a nutrient flush associated with rewetting dry soil was minimised (e.g. Fierer and Schimel 2002) and treatments were imposed.
Treatments with and without N (25 kg [ha.sup.-1]) were applied to the soil and foliage were in the form of REE (La, Ce, Y, Ho, Tb, Yb, and La and Ce combined) chlorides applied at 0.5 and 2.0 kg [ha.sup.-1], except for the La and Ce mixture which was applied at 0.5 kg La [ha.sup.-1] plus 0.5 kg Ce [ha.sup.-1]. These REEs were chosen due to their lower cost and easier availability compared with other REEs. There were eight replicates per treatment. Each REE (and N) treatment was applied as a spray (50 mL per pot): cited by Guo (1993) as the most common application method for crops such as barley, cotton, maize, rapeseed, rice, ryegrass, and sugar beet. Seven harvests were taken from the trial on a monthly basis. The plants were trimmed with hand shears to ~30-40 mm height and the foliage dried at 65[degrees]C for 48 h and weighed. A single bulk sample for each pot was made across harvests by combining all DM for a specific pot. The bulk samples were ground to <1 mm and analysed for La, Ce, Y, Ho, Tb, Yb and Cd concentration by ICP-OES following a nitric acid microwave digest (Zhang and Shan 1997). A sub-set of 10 foliage samples, across the range of concentrations measured, were also digested in a nitric acid-hydrogen peroxide mixture and measured by ICP-OES (Spalla et al. 2009). Concentrations were not statistically different to concentrations measured after a nitric acid digest. Hence, a decision was made to digest samples using the quicker and cheaper method of Zhang and Shan (1997).
Additional glasshouse testing
Following glasshouse screening, La was chosen for additional testing in three different soils and moisture regimes. The soils were a Horotiu silt loam (USDA Classification: Typic Udivitrand), a Lismore stony silt loam (considered analogous to the Warepa soil, which was unavailable; USDA Classification: Udic Haplustept) and a Conroy fine sandy loam (USDA Classification: Aridic Haplustalf), representative of the Allophanic, Pallic, and Semiarid New Zealand Soil Orders, respectively. Samples of each soil were obtained and analysed in the same manner as the initial glasshouse screening. Mean chemical and physical characteristics of each soil are given in Table 1.
For each soil type, ~3 kg of air-dried soil was put into each pot and adjusted to 20, 40 or 70% ASM. Wilting point for the Horotiu, Lismore and Conroy soils was, 23, 18 and 19% v/v, respectively; while the corresponding field capacity values were 48, 38 and 34% v/v. Soils were maintained at these moistures by weighing a subsample of pots every few days and watering if necessary. A row of buffer-pots were arranged around the outside of the trial to minimise edge effects. Ryegrass was treated as per the initial screening experiment for 9 weeks. One month after initialisation the first harvest was taken. The La treatments were then sprayed onto each soil by moisture treatment at a rate of 0,0.5,1.0,5 or 10 kg La [ha.sup.-1], each with five replicates (i.e. 75 pots per soil). Foliage was harvested three times before another La application was made at the same rates and three more harvests made, giving a total of seven harvests and two REE applications. Treatments are referred to as the sum of La applied (e.g. the greatest rate was 20 kg La [ha.sup.-1]).
Following a preliminary analysis of the first three harvests, data suggested that there was a yield effect due to La application. Unfortunately, we could not determine if this also resulted in different chemical composition (e.g. less Cd in foliage) in a bulked sample due to the loss of the second harvest. Foliage from the control and 20 kg La [ha.sup.-1] treatments of the first harvest after the second application of La was dried, ground <1 mm, acid digested (Kalra et al. 1989) and analysed via ICP-OES for total N, P, K and Cd concentrations.
A field trial sites was established on the AgResearch Invermay research farm (Warepa silt loam: USDA Classification; Udic Haplustept) near Dunedin, New Zealand. The site supported a mixed clover and ryegrass pasture grazed by sheep and received regular dressings of superphosphate (10 kg P [ha.sup.-1] [year.sup.-1]) and limestone (1 Mg [ha.sup.-1] every 3 4 years). The trial began in spring, 2011 and ran for 2 years.
The site was fenced to exclude stock and DM harvested and removed to a uniform height of 5 cm above the soil surface (~1300 kg [ha.sup.-1] residual cover). Forty eight plots (2 m x 1 m with a 1 m buffer area between) were established. A calibration cut was carried out initially to help standardise natural variation between plots.
The plots were arranged in a grid, with six replicates each of La applied at zero, 0.5 kg [ha.sup.-1], 0.5 kg La [ha.sup.-1] plus 0.5kg Ce [ha.sup.-1], or 2 kg La [ha.sup.-1], all with and without N applied at 25 kg [ha.sup.-1] (as urea). Treatments, designed to be consistent with the glasshouse trial, were applied in spring and late summer using analytical grade La[Cl.sub.3] x 7[H.sub.2]0, Ce[Cl.sub.3] x 7[H.sub.2]0, and urea dissolved into 1 L of deionised water then sprayed onto each plot ensuring even application to foliage and soil surface. Application took place 10-14 days after harvest dates to allow for sufficient fresh foliage to uptake the REEs and/or N.
Seven harvests were taken at approximately monthly intervals during the growing season (Sept.-May), using a rotary mower set at a height of 5 cm above the soil surface (to leave residual dry matter of ~1300 kg [ha.sup.-1]). Foliage was weighed and a 200-300 g representative sub-sample dried at 65[degrees]C for 48 h and re-weighed to estimate DM yield. In addition to a sample bulked across all harvests, individual samples for harvests taken two and seven weeks after the first REE application were also ground <1 mm, nitric acid microwave digested (Zhang and Shan 1997) and analysed via ICP-OES for total Cd.
All data was log-transformed before analysis if not normally distributed before being analysed by ANOVA and checking residuals for homoscedasticity.
For the soil survey, point-wise data for each REE were analysed by ANOVA, fitting terms for the factorial interaction of soil parent material and land use.
For the initial glasshouse screening, cumulative yield per pot after REE application DM and mean Cd concentrations were analysed by ANOVA, fitting terms for the factorial interaction of REE type, rate and N application. After the isolation of a N effect, treatments were compared using a Duncan's multiple range test.
For the additional glasshouse testing, mean DM yield for all harvests and Cd concentration in foliage for the harvest following the second La application were analysed by ANOVA fitting terms for the factorial interaction of soil type and soil moisture with La application. For simplicity, a comparison (i-test) of Cd concentrations was made at 0 and 20 (i.e. 2 x 10 kg [ha.sup.-1]) application rates.
For the field trial an ANOVA was then used to compare cumulative yield between treatments and Cd concentrations in selected harvests after REE application, with comparisons made between treatments and years with a Duncan's multiple range test. A calibration cut was used as a covariate in the analysis of data to account for inherent differences in DM between the plots.
Data for the mean concentrations of REEs and Cd in soils and fertilisers are given in Table 2. The concentrations of many REEs were greater in Volcanic compared with Sedimentary soils. A greater concentration of REEs in Peat soils compared with Sedimentary soils reflected the lower bulk density of the Peat soils. Dairy soils were also more enriched in certain REEs (Ce, Eu, La and Nd) compared with Sheep and Beef soils. This enrichment may have been caused by the greater application of P-fertilisers to dairy soils, since the same four REEs were also enriched relative to other elements in the P-fertilisers analysed. Zhang et al. (2006) estimated that the application of 320 kg [ha.sup.-1] [year.sup.-1] superphosphate manufactured in Zhinjin County, China, would supply between 38-189 g REE [ha.sup.-1] annually to plants.
The concentration of REEs in soil is also influenced by parent material decreasing from: granite > quaternary > basalt > sandstone (Hu et al. 2006). Loessial and calcareous soils tend to have lower total REE concentrations (137-174 mg [kg.sup.-1]) than those derived from igneous rock or sandstone (174-219 mg [kg.sup.-1]). Surveys have thus far shown the greatest enrichment of REEs in soils from China, which had a mean total REE concentration of 177 mg [kg.sup.-1], whereas studies of soils in other countries put mean concentrations at between 18 and 118 mg [kg.sup.-1] (Hu et al. 2006). The range of parent materials in our (New Zealand) survey resulted in a mean total concentration of REEs 67 mg [kg.sup.-1], greater than soils from Germany or the USA, but less than soils from Australia, Sweden, Japan or Malaysia. We therefore conclude that the mean REE concentration in New Zealand soils is no more enriched than those from many other countries, but localised enrichment due to soil order or land use may occur.
The addition of N significantly (P< 0.001) increased overall DM yield by 32% (Table 3). Additional analysis of DM in REE treatments either with or without N additions showed that there were both increases and decreases in yield compared with the control. However, Ho decreased yield by >5% irrespective of N additions while a similar decrease occurred for Y addition in conjunction with N. Hence, both were discounted from further investigation. Interestingly, of the REEs that remained--La and Ce; both are included in commercial REE fertilisers available from China (e.g. Changle[TM] contains 25% Ce[0.sub.2] and 12% La[0.sub.3]; Xiong et al. 2000). After considering the availability and price of the REEs that remained (Arafura Resources 2013), La and Ce were selected as readily available and of low enough cost to warrant further testing.
Application of N had no effect on Cd concentrations (Table 3). The addition of La averaged across the 2.0 kg [ha.sup.-1] application rates (irrespective of N addition) decreased Cd foliage concentration by -18% compared with the control (Table 3). The effects of adding other REEs were not significant. For its potential yield and effect on decreasing Cd uptake, La was the focus of additional glasshouse and field testing.
Additional glasshouse testing
Overall DM yield was affected by the rate of La applied in the Conroy and Lismore soils (Fig. 1), but not in the Horotiu soil perhaps because of a greater REE concentration in the Volcanicderived Horotiu soil compared with the Sedimentary-derived Conroy and Lismore soils. When broken into individual harvests, overall yield was largely accounted for by increases for the two harvests after La was applied to the Conroy soil and for the first harvest after application for the Lismore soil (Fig. 2).
Data from the first harvest after the second application of La showed that Cd concentrations in foliage paralleled soil Cd concentrations, greatest in the Horotiu soil and least in the Conroy soil (Tables 1 and 4). On average, concentrations of Cd in foliage decreased in the 20 kg La [ha.sup.-1] treatment compared with the control by 60, 40 and 30% in the Conroy, Lismore and Horotiu soils, respectively.
Summarising the effects of treatments in the first year, the addition of N increased DM yield over the control by 9%, 17% when N was added with 0.5 kg La [ha.sup.-1], and 4% when only 0.5 kg La [ha.sup.-1] was added (Table 5). The addition of 2 kg La [ha.sup.-1] with N increased yield by -10%, but decreased yield without N added. No other treatments were significant. During the second year only treatments with N added improved yield over the control (Table 5). Notably, mean annual temperature during year one was close to the 30 year average, but much cooler in the second year (annual temperature 1.3[degrees]C below the 30 year average).
Foliage composition data for a sample of all harvests and both years showed that mean N, P and S concentrations were consistent across the treatments and sites, but K concentrations were slightly enriched for the 0.5 kg La [ha.sup.-1] and 0.5 kg La+ 0.05 kg Ce [ha.sup.-1] treatments (Table 6). The concentration of Cd determined in foliage taken in the first harvest two weeks after application showed more significant decreases relative to the control compared with harvests taken seven weeks after the first and second treatment application. However, there was no overall effect on Cd concentration, suggesting that any decrease in Cd enrichment of foliage was short lived.
Effect on yield and Cd concentrations in foliage
Both glasshouse and field trials indicated mixed results with a modest increase in DM yield following the application of La, but only under specific conditions of low soil REE concentration and soil moisture. The DM responses to La were consistent with the little literature that exists on the impact of REE. For example, both the Conroy and Lismore soils contained lower concentrations of La and Ce than the Horotiu soil and hence were more likely to exhibit La response. The Conroy and Lismore soils also had less ASM (14% v/v and 20% v/v, respectively) than the Horotiu soil (25% v/v), and were therefore more likely to promote plant water stress--a condition Buckingham et al. (1999) attributed as a likely cause for summer yield improvements of 24% for ryegrass on a sandy soil in Victoria, Australia. Similarly, increases in K concentrations when treated with La or Ce were similar to those exhibited in rice (12 and 9% for P and K, respectively; Ning and Xiao 1989) and wheat (8% for P; Zhu et al. 1994). However, this is probably due to the stimulation of K uptake caused by the addition of REEs as chlorides rather than a REE effect (Bloom and Finazzo 1986). This is supported by the fact that all three soils had P and K concentrations considered greater than that required for optimal pasture growth (Edmeades et al. 2010).
Soil chemical properties may have also influenced Cd concentrations in foliage. For instance, mean concentrations in foliage paralleled the Cd concentrations and pH (which increases Cd uptake with increasing acidity) among the different soils. However, there are also reports indicating that REEs can decrease Cd concentrations when sprayed onto foliage, irrespective of soil type (Liu et al. 1997; Wang et al. 2006), although the mechanism causing the decrease in Cd concentration is unclear.
There are several reports showing an interaction between Cd uptake and N applications. For instance, Li et al. (2011) reported a 64% increase in Cd uptake for wheat when N was applied. However, both increases and decreases have been noted. Generally, a dilution effect is ascribed to plant growth quicker than the rate of Cd desorption from solid to liquid phase into soil solution (Landberg and Greger 2003; Wangstrand et al. 2007), whereas enrichment has been noted where Cd supply is good and conditions are favourable for growth.
Climatic conditions in the glasshouse studies were controlled. However, in the field, cool temperatures (annual temperature 1.3[degrees]C below the 30 year average) may have contributed to a lack of a REE effect in the second year. Similarly, periods of water stress which yielded improvements in ryegrass yield in Victoria, Australia (Buckingham et al. 1999) were rare for the months following treatment applications. Clearly, more work is required to establish any mechanism of action, but even in a soil where the effect of adding REEs is likely, yield and Cd uptake responses were modest and short-lived.
The commonly available REEs are Ce and La chloride or nitrate salts. These materials are highly soluble but crystalline in nature. Application of the pure salts would require the use of spray application as the crystals are not suitable for broadcast or spin application to fields. In addition, consistent application across a field will be problematic due to the low application rates of 0.5 kg REE [ha.sup.-1]. The most convenient method for application of REEs is utilising their compatibility with a urea-N solution which can be spray applied or delivered via fertigation to ensure even coverage to the target crop or pasture.
Rare earth elements are expensive relative to other agronomically important nutrients such as N, P or S. The high relative cost is not due to the technology to extract the elements, but mostly due to the international trade and market conditions, which are dominated by China. Since at least 2009, China has had export restrictions on REE exports with exporters subject to quotas on supply (Hurst 2010). As a consequence, REE prices have been volatile over this period with the La oxide price exceeding $US60[kg.sup.-1] (Arafura Resources 2013). With prices consistently over $US10 [kg.sup.-1] for La oxide since 2009, a large yield response would need to occur to offset costs. The results showed that the widespread use of REEs such as La is uneconomic due to the narrow range environmental conditions and soils with low enough REE concentrations to possibly show a yield response. Environmentally, the inhibition of Cd uptake was also only short-term. Moreover, repeated applications of REEs may lead to their accumulation in the soil to concentrations that may become problematic, such as for the germination or growth of some crops (Thomas et al. 2014).
Soil survey data indicated that mean concentrations of many REEs were enriched in Volcanic compared with Sedimentary soils and for soils used for dairying compared with sheep and beef. Land use differences were attributed to factors such as inputs by REE-rich P-fertilisers. From a range of REEs, La and Ce showed some potential to increase DM yield of ryegrass. However, additional glasshouse testing only showed a yield improvement for two soils with low La and Ce concentration that received low rates of La and were maintained at 40% ASM or less. A mixed response was also evident in the field, with the addition of La only boosting yield at low rates (0.5 kg [ha.sup.-1]) and in one year when temperatures and soil moisture were near the 30 year average. Applying La caused a decrease in Cd concentrations in foliage for the first harvest after application, but did not result in a significant decrease over all harvests. Hence, we conclude that there was insufficient data to warrant the use of La (screened from a range of REEs) as a strategy to increase dry matter yield and decrease Cd concentrations in ryegrass-dominated pastures in New Zealand.
This work was supported by the Ballance Agri-Nutrients' Primary Growth Partnership with the Ministry for Primary Industries entitled 'Clearview'.
Arafura Resources (2013) Arafura Resources, Rare Earth Elements Pricing. Available at: www.arafuraresources.com.au/rare-earths/pricing (accessed Sept. 2015).
Blakemore LC, Searle PL, Daly BK (1987) 'Methods for chemical analysis of soils.' p. 103. New Zealand Soil Bureau Scientific Report 80. (NZSB: Wellington, New Zealand)
Bloom AJ, Finazzo J (1986) The influence of ammonium and chloride on potassium and nitrate absorption by barley roots depends on time of exposure and cultivar. Plant Physiology 81, 67-69. doi: 10.1104/pp.81.1.67
Boger M, Van Erp YHM, Eichner P (1997) Assessment of the potential toxicity of lanthanum for aquatic organisms. In 'Proceedings of the 2nd Sino-Dutch Workshop on the Environmental Behavior and Ecotoxicology of Rare Earth Elements and Heavy Metals'. (Eds R Guicherit, W Zhu) pp. 14-18. (TNO Publication: La Delft, the Netherlands)
Brioschi L, Steinmann M, Lucot E, Pierret MC, Stifle P, Prunier J, Badot PM (2013) Transfer of rare earth elements REE from natural soil to plant systems, implications for the environmental availability of anthropogenic REE. Plant and Soil 366, 143-163. doi: 10.1007/s 11104-012-1407-0
Buckingham S, Maheswaran J, Meehan B, Peverill K. (1999) The role of applications of rare earth elements in enhancement of crop and pasture production. Materials Science Forum 315-317, 339-347. doi:10.4028/www.scientific.net/MSF.315-317.339
Cadmium Working Group (2011) Cadmium and New Zealand agriculture and horticulture: A strategy for long term risk management. A report prepared by the Cadmium Working Group for the Chief Executives Environmental Forum. Available at: http://gdsindexnz.org/wp-content/uploads/2015/02/ GDS031.-Cadmium-and-New-Zealand-Agricultureand-Horticulture-a-Strategy-for-Long-Term-Risk-Management -2011.pdf (accessed 24 August 2015).
Drewry JJ, Littlejohn RP, Paton RJ (2000) A survey of soil physical properties on sheep and dairy farms in southern New Zealand. New Zealand Journal of Agricultural Research 43, 251-258. doi: 10.1080/00288233.2000.9513425
Edmeades DC, Morton JD, Waller J, Metherell AK, Roberts AHC, Carey PJ (2010) The diagnosis and correction of potassium deficiency in New Zealand pastures: a review. New Zealand Journal of Agricultural Research 53, 151-173. doi: 10.1080/00288233.2010.482954
Fierer N, Schimel JP (2002) Effects of drying-rewetting frequency on soil carbon and nitrogen transformations. Soil Biology & Biochemistry 34, 777-787. doi: 10.1016/S0038-0717(02)00007-X
Grewal KS, Buchan GD, Tonkin PJ (1990) Estimation of field capacity and wilting point of some New Zealand soils from their saturation percentages. New Zealand Journal of Agricultural Research 18, 241-246.
Guo BS (1993) The application of rare earth elements on agriculture and breeding. Journal of the Chinese Rare Earths Society 15, 37-43.
Guo BS, Zhu WM, Xiong PK, Ji YJ, Liu Z, Wu ZM (1988) 'Rare earths in agriculture.' pp. 23-208. (Agricultural Scientific Technological Press: Beijing, China) [in Chinese]
Hu Z, Haneklaus S, Sparovek G, Schung E (2006) Rare earth elements in soils. Communications in Soil Science and Plant Analysis 37, 1381-1420. doi: 10.1080/00103620600628680
Hurst C (2010) China's rare earth elements industry, what can the west learn? Institute for the Analysis of Global Security. Available at: http://americanresources.org/wp-content/uploads/2011 /09/rareearth.pdf (accessed Sept 2015).
Kalra YP, Maynard DG, Radford FG (1989) Microwave digestion of tree foliage for multi-element analysis. Canadian Journal of Forest Research 19, 981-985. doi:10.1139/x89-150
Landberg T, Greger M (2003) Influence of N and N supplementation on Cd accumulation in wheat grain. In 'Proceedings of 7th International Conference on the Biogeochemistry of Trace Elements 1, III'. Uppsala, Sweden, pp. 90-91.
Li X, Ziadi N, Belanger G, Cai Z, Xu H (2011) Cadmium accumulation in wheat grain as affected by mineral N fertilizer and soil characteristics. Canadian Journal of Soil Science 91, 521-531. doi: 10.4141/cjss 10061
Liang T, Zhang S, Wang L, Rung H-T, Wang Y, Hi A, Ding S (2005) Environmental biogeochemical behaviors of rare earth elements in soilplant systems. Environmental Geochemistry and Health 27, 301-311. doi: 10.1007/s10653-004-5734-9
Liu HZ, Wen LK, Li YD, Rui H (1997) Effects of rare earth on the growth, fruiting and absorption of N, P, K on black currant. Journal of Fruit Science 14, 32-35.
Liu X, Wang J, Yang J, Fan Y, Wu Y, Zhang H (2006) Application of rare earth phosphate fertilizer in western area of China. Journal of Rare Earths 24, 423-426. doi: 10.1016/S1002-0721 (07)60418-9
McDowell RW, Taylor MD, Stevenson BA (2013) Natural background and anthropogenic contributions of cadmium to New Zealand soils. Agriculture, Ecosystems & Environment 165, 80-87. doi: 10.1016/j.agee.2012.12.011
Ning JB, Xiao SL (1989) Effects of rare earth elements application on day lily. Chinese Rare Earths 10, 52-54.
Otero N, Vitoria L, Soler A, Canals A (2005) Fertiliser characterization, major, trace and rare earth elements. Applied Geochemistry 20, 1473-1488. doi: 10.1016/j.apgeochem.2005.04.002
Rambeck WA, Wehr U (2005) Rare earth elements as alternative growth promoters in pig production. Animal Science Database. Available at: www.lanthanoide.de/fileadmin/docs_archiv/Bilder/REE_pig_review.pdf (accessed Sept 2015).
Spalla S, Bafifi C, Barbante C, Turetta C, Cozzi G, Beono GM, Bettinelli M (2009) Determination of rare earth elements in tomato plants by inductively coupled plasma mass spectroscopy techniques. Rapid Communications in Mass Spectrometry 23, 3285-3292. doi:10.1002/rcm.4244
Thomas PJ, Carpenter D, Boutin C, Allison JE (2014) Rare earth elements (REEs): Effects on germination and growth of selected crop and native plant species. Chemosphere 96, 57-66. doi:10.1016/j.chemosphere. 2013.07.020
Tyler G (2004) Rare earth elements in soil and plant systems - a review. Plant and Soil 267, 191-206. doi: 10.1007/s 11104-005-4888-2
Wald PH (1990) 'A review of the literature on the toxicity of rare earth metals as it pertains to the engineering demonstration system surrogate testing.' pp. 1-27. (Lawrence Berkeley National Laboratory: CA; Dept of Energy: Washington, DC)
Wang Z, Liu D, Lu P, Wang C (2001) Accumulation of rare earth elements in com after agricultural application. Journal of Environmental Quality 30, 37-45. doi: 10.2134/jeq2001.30137x
Wang J, Liu X, Yang J, Zhang H, Liu Y, Fan Y, Wu Y, Han X (2006) Development and prospect of rare earth functional biomaterials for agriculture in China. Journal of Rare Earths 24, 427-431. doi : 10.1016/S 1002-0721 (07)60419-0
Wangstrand H, Eriksson J, Oborn I (2007) Cadmium concentration in winter wheat as affected by nitrogen fertilization. European Journal of Agronomy 26, 209-214. doi:10.1016/j.eja.2006.09.010
Xie ZB, Zhu JG, Chu HY, Zhang YL, Zeng Q, Ma HL, Cao ZH (2002) Effect of lanthanum on rice production, nutrient uptake, and distribution. Journal of Plant Nutrition 25,2315-2331. doi: 10.1081 /PLN-120014078
Xiong BK, Cheng P, Guo BS, Zheng W (2000) 'Rare earth element research and applications in Chinese agriculture and forest.' (Metallurgical Industry Press: Beijing, China)
Xu X, Wang Z (2007) Phosphorus uptake and translocation in field-grown maize after application of rare earth-containing fertilizer. Journal of Plant Nutrition 30, 557-568. doi:10.1080/01904160701209287
Zhang SZ, Shan XQ (1997) The determination of rare earth elements in soil by inductively coupled plasma mass spectrometry. Atomic Spectroscopy 18, 140-144.
Zhang Y, Deng L-B, Ye W-H (2006) Does the long-term application of calcium superphosphate lead to an increase of the soil rare earth element contents? Journal of Environmental Sciences 18, 130-134.
Zhu YY, Chen JJ, Yang JC, Liu XL, Liu LH (1994) Effect of rare earth with high-grade La on the yield, uptake of nitrogen and phosphoms by spring wheat. Acta Agriculturae Nucleatae Sinica 8, 46-50.
Zhu JG, Chu HY, Xie ZB, Yagi K (2002) Effects of lanthanum on nitrification and ammonification in three Chinese soils. Nutrient Cycling in Agroecosystems 63, 309-314. doi: 10.1023/A: 102116310 1692
R. W. McDowell (A,B,D), W. Catto (C), and T. Orchiston (A)
(A) AgResearch, Invermay Agricultural Centre, Private Bag 50034 Mosgiel, New Zealand.
(B) Faculty of Agriculture and Life Sciences, Lincoln University. Lincoln 7647, New Zealand.
(C) Ballance Agri-Nutrients, Private Bag 12503, Tauranga, New Zealand.
(D) Corresponding author. Email: firstname.lastname@example.org
Table 1. Mean chemical and physical characteristics of soil (New Zealand soil order in parentheses) used in the initial screening, additional glasshouse testing and the field trial Parameter Initial Horotiu Additional Conroy scrcening/ (Allophanic) glasshouse (Semi- Field testing arid) trial Lismore Warepa (Pallie) (Pallie) ASC (%) 24 84 26 7 CEC (me 100 14 20 15 8 [g.sup.-1]) Bulk density 1.03 0.88 1.02 1.02 (g [cm.sup.-3]) Olsen P 7 28 10 25 (mg [L.sup.-1]) pH 5.9 5.1 5.7 7.1 Quick test K (A) 6 8 8 6 Total Cd 0.25 0.70 0.07 0.05 (mg [kg.sup.-1]) Total C 37 57 31 14 (g [kg.sup.-1]) Total N 3.1 6.3 3.0 1.7 (g [kg.sup.-1]) (A) Quick test unit = 20.8 x bulk density (g [cm.sup.-3]) x exchangeable K (me 100 [g.sup.-1]). Table 2. Mean concentration (mg [kg.sup.-1]) of Cd and rare earth elements in soils and fertilisers by land use and soil parent material The F-statistic is also given for the significance of treatment comparisons by land use, soil parent material and their interaction. Enrichment ratio is the quotient of overall mean P-fertiliser and soil mean concentrations for each REE. Shaded cells indicate those REEs most enriched, and thereby indicative of, P-fertilisers Soil parent material Cd Ce Dy Er Peat 1.43 9.3 2.2 4.2 Sedimentary 0.77 8.2 3.1 3.3 Volcanic 0.99 26.2 4.9 6.6 Sedimentary 0.90 8.6 3.6 4.2 Volcanic 1.01 18.6 4.2 4.6 Land use 0.455 0.036 0.691 0.294 Soil <0.001 <0.001 <0.001 <0.001 Interaction 0.427 0.027 0.175 <0.001 P-fertilisers -- 76.5 13.4 13.6 Overall soils mean -- 13.8 3.7 4.5 Enrichment ratio -- 5.6 3.6 3.0 Soil parent material Eu Gd Ho La Peat 13.3 0.9 0.2 4.8 Sedimentary 6.6 0.6 0.1 3.2 Volcanic 4.2 2.4 0.4 23.2 Sedimentary 5.5 0.7 0.1 3.9 Volcanic 2.6 1.5 0.3 16.2 Land use <0.001 0.052 0.078 0.085 Soil <0.001 <0.001 <0.001 <0.001 Interaction 0.707 0.027 0.047 0.066 P-fertilisers 130.2 5.2 1.2 52.1 Overall soils mean 5.7 1.2 0.2 9.7 Enrichment ratio 22.8 4.5 5.2 5.4 Soil parent material Lu Nd Pr Sm Peat 0.1 36.9 1.1 1.1 Sedimentary 0.3 11.8 1.2 1.8 Volcanic 0.4 29.8 4.8 3.9 Sedimentary 0.4 15.1 1.4 2.1 Volcanic 0.4 13.8 3.8 3.4 Land use 0.024 0.002 0.298 0.745 Soil <0.001 <0.001 <0.001 <0.001 Interaction 0.694 0.001 0.137 0.196 P-fertilisers 0.8 445.4 6.9 5.7 Overall soils mean 0.4 19.1 2.4 2.5 Enrichment ratio 2.1 23.3 2.8 2.3 Soil parent material Tb Th Tm Yb Peat 0.2 0.1 0.2 0.5 Sedimentary 0.5 0.5 0.4 0.4 Volcanic 0.7 0.2 0.6 1.2 Sedimentary 0.5 0.1 0.5 0.4 Volcanic 0.6 0.8 0.8 0.8 Land use 0.158 0.496 0.173 0.075 Soil <0.001 0.110 0.008 <0.001 Interaction 0.365 <0.001 0.677 0.034 P-fertilisers 0.9 3.4 0.5 4.0 Overall soils mean 0.5 0.3 0.5 0.6 Enrichment ratio 1.6 10.4 0.9 6.2 Table 3. Mean dry matter (DM) total yield (mg [pot.sup.-1]) and Cd concentration (mg [kg.sup.-1]) of foliage from the initial glasshouse screening pot trial Treatments for yield or Cd concentration are compared within each N application rate using Duncan's multiple range test at the P = 0.05 level of significance. Within each row, means followed by the same letter are not significantly different Treatment 0kg N [ha.sup-1] 25 kg N [ha.sup.-1] DM yield Cd cone. DM yield Cd cone Control 3094a 0.175a 3874a 0.205a 0.5 kg [ha.sup.-1] Ce 3259b 0.179a 3706a 0.183a 2 kg [ha.sup.-1] Ce 3468c 0.164a 3812a 0.179a 0.5 kg [ha.sup.-1] Ho 2732b 0.199a 3802a 0.193a 2 kg [ha.sup.-1] Ho 2788b 0.196a 3847a 0.202a 0.5 kg [ha.sup.-1] La 3194a 0.149b 3777a 0.176a 2 kg [ha.sup.-1] La 3346c 0.143b 4153b 0.165b 0.5 kg [ha.sup.-1] Tb 3173a 0.166a 3877a 0.173a 2 kg [ha.sup.-1] Tb 3135a 0.174a 4043a 0.169a 0.5 kg [ha.sup.-1] Yb 3065a 0.199a 3985a 0.172a 2 kg [ha.sup.-1] Yb 3106a 0.164a 3987a 0.194a 0.5 kg [ha.sup.-1] Y 3102a 0.178a 3163c 0.181a 2 kg [ha.sup.-1] Y 3353c 0.163a 3670c 0.184a 0.5 kg ha 1 La + 0.5 3177a 0.145b 3938a 0.178a kg [ha.sup.-1] Ce Table 4. Mean Cd concentration in foliage from the first harvest after the second application of La in the additional glasshouse trial The F-statistic for the significance of treatment comparisons is given for soil type, La application rate, soil moisture and their interactions. An asterisk indicates a significant contrast between each La by soil type by soil moisture treatment and the control using a pairwise t-test Soil type La rate (kg Soil Cd [ha.sup.-1]) moisture (mg [kg.sup.-1]) (% v/v) Conroy (Semi-arid) 0 20 0.171 -- 40 0.060 70 0.048 20 20 0.040 * 40 0.031 * 70 0.042 Lismore (Pallie) 0 20 0.127 40 0.289 70 0.090 20 20 0.077 * 40 0.152 * 70 0.067 Horotiu (Volcanic) 0 20 0.338 40 0.343 70 0.284 20 20 0.302 -- 40 0.253 * -- 70 0.223 Soil <0.001 La rate 0.002 Moisture 0.023 Soil x La rate 0.923 Soil x moisture 0.028 La rate x moisture 0.501 Soil x La rate x 0.449 moisture Table 5. Mean dry matter yield (kg [ha.sup.-1]) for two years in the field trial at Invermay Mean yields for each year are compared between treatments using Duncan's multiple range test at the P=0.05 level of significance. Within each row, means followed by the same letter are not significantly different Treatment Year 1 Year 2 Control 8090a 6000a 25 kg [ha.sup.-1] N 8820b 6700b 0.5 kg [ha.sup.-1] La 8500b 6350a 0.5 kg [ha.sup.-1] La+ 25 kg [ha.sup.-1] N 9450c 7150b 0.5 kg [ha.sup.-1] La + 0.5 kg [ha.sup.-1] Ce 7780a 5950a 0.5 kg [ha.sup.-1] La + 0.5 kg [ha.sup.-1] 9190b 6950b Ce + 25 kg N 2 kg [ha.sup.-1] La 7660a 6100a 2 kg [ha.sup.-1] La + 25 kg Ce+ 25 kg N N 8930b 7100b Table 6. Mean N, P, K, S and Cd concentration in a bulked sample of foliage from all harvests taken from the field trial, and the mean Cd concentration in foliage sampled from the first harvest after the first REE application in year I and the second REE application in year 2 Mean concentrations are compared between treatments using Duncan's multiple range test at the P 0.05 level of significance. Within each row, means followed by the same letter are not significantly different Treatment N P (g [kg.sup.-1]) (g [kg.sup.-1]) Control 29.2a 4.1a 25 kg [ha.sup.-1] N 29.9a 4.2a 0.5 kg [ha.sup.-1] La 29.2a 4.3a 0.5 kg [ha.sup.-1] 29.8a 4.2a La + 25 kg [ha.sup.-1] N 0.5 kg [ha.sup.-1] 28.7a 4.2a La + 0.5 kg [ha.sup.-1] Ce 0.5 kg [ha.sup.-1] 29.4a 4.2a La + 0.5 kg [ha.sup.-1] Ce + 25 kg N 2 kg [ha.sup.-1] La 28.3a 4.3a 2 kg [ha.sup.-1] 29.9a 4.2a La + 25 kg [ha.sup.-1] N Treatment K S (g [kg.sup.-1]) (g [kg.sup.-1]) Control 16.1a 4.4a 25 kg [ha.sup.-1] N 15.5a 4.2a 0.5 kg [ha.sup.-1] La 17.6b 4.5a 0.5 kg [ha.sup.-1] 16.5a 4.3a La + 25 kg [ha.sup.-1] N 0.5 kg [ha.sup.-1] 16.9a 4.5a La + 0.5 kg [ha.sup.-1] Ce 0.5 kg [ha.sup.-1] 18.0b 4.3a La + 0.5 kg [ha.sup.-1] Ce + 25 kg N 2 kg [ha.sup.-1] La 17.0a 4.6a 2 kg [ha.sup.-1] 16.6a 4.3a La + 25 kg [ha.sup.-1] N Treatment Cd Concentration of (mg [kg.sup.-1]) Cd (mg [kg.sup.-1]) 2 weeks after 1st application Control 0.110a 0.190a 25 kg [ha.sup.-1] N 0.122a 0.186a 0.5 kg [ha.sup.-1] La 0.081a 0.110b 0.5 kg [ha.sup.-1] 0.104a 0.127b La + 25 kg [ha.sup.-1] N 0.5 kg [ha.sup.-1] 0.122a 0.137a,b La + 0.5 kg [ha.sup.-1] Ce 0.5 kg [ha.sup.-1] 0.093a 0.117b La + 0.5 kg [ha.sup.-1] Ce + 25 kg N 2 kg [ha.sup.-1] La 0.106a 0.135b 2 kg [ha.sup.-1] 0.109a 0.114b La + 25 kg [ha.sup.-1] N Treatment Concentration of Cd (mg [kg.sup.-1]) 7 weeks after 1st application Control 0.185a 25 kg [ha.sup.-1] N 0.177a 0.5 kg [ha.sup.-1] La 0.155a 0.5 kg [ha.sup.-1] 0.175a La + 25 kg [ha.sup.-1] N 0.5 kg [ha.sup.-1] 0.155a La + 0.5 kg [ha.sup.-1] Ce 0.5 kg [ha.sup.-1] 0.169a La + 0.5 kg [ha.sup.-1] Ce + 25 kg N 2 kg [ha.sup.-1] La 0.171a 2 kg [ha.sup.-1] 0.148a La + 25 kg [ha.sup.-1] N
|Printer friendly Cite/link Email Feedback|
|Author:||McDowell, R.W.; Catto, W.; Orchiston, T.|
|Date:||Oct 1, 2015|
|Previous Article:||Spatial and temporal distribution of rainfall erosivity in New Zealand.|