Cadmium accumulation by willow clones used for soil conservation, stock fodder, and phytoremediation.
Abstract: Elevated levels of cadmium are often found in the soil of New Zealand pasturelands due to the long-term use of Cd-contaminated fertilisers. The accumulation of Cd in willow biomass used as stock fodder could therefore adversely affect agricultural productivity and human health. Alternatively, willows may be used for phytoremediation of Cd-contaminated soil at polluted sites. An investigation was carried out to determine the variation in Cd as well as Zn, Mn, and Fe accumulation in 15 willow clones that had been bred for soil conservation purposes. These clones were grown under controlled conditions in 20-L pots of soil containing Cd, Zn, Mn, and Fe at concentrations of 0.3, 64, 597, and 56000 mg/kg, respectively. Daily water use was measured over the final 2 weeks of the experiment and biomass accumulation was determined at the end of the experiment. We found that shrub willows had significantly higher leaf and stem Cd, Mn, and Zn concentrations than tree willows. Average leaf Cd concentrations varied widely between clones from 1.5 to 10 mg/kg. Clones with a high Cd accumulation capacity may be selected to improve the efficacy of Cd-phytoremediation, whereas clones that accumulated lower Cd concentrations may be used for stock fodder. Metal concentrations were not significantly correlated with plant water-use, or biomass production.

Additional keywords: bioaccumulation, heavy metal, pastureland, water use.
Subject: Pasture ecology (Research)
Cadmium (Environmental aspects)
Willows (Environmental aspects)
Bioaccumulation (Research)
Authors: Granel, Thierry
Robinson, Brett
Mills, Tessa
Clothier, Brent
Green, Steve
Fung, Lindsay
Pub Date: 12/15/2002
Publication: Name: Australian Journal of Soil Research Publisher: CSIRO Publishing Audience: Academic Format: Magazine/Journal Subject: Agricultural industry; Earth sciences Copyright: COPYRIGHT 2002 CSIRO Publishing ISSN: 0004-9573
Issue: Date: Dec 15, 2002 Source Volume: 40 Source Issue: 8
Topic: Event Code: 310 Science & research
Product: Product Code: 3339640 Cadmium NAICS Code: 331419 Primary Smelting and Refining of Nonferrous Metal (except Copper and Aluminum) SIC Code: 3339 Primary nonferrous metals, not elsewhere classified
Geographic: Geographic Scope: New Zealand Geographic Code: 8NEWZ New Zealand
Accession Number: 96237877
Full Text: Introduction

Willows are used extensively in New Zealand for soil conservation and supplementary stock fodder during times of drought (Wilkinson et al. 1999). Both foliage and small twigs can be browsed by sheep and cattle (Hathaway 1986; Douglas et al. 1996). The majority of New Zealand pasturelands have elevated cadmium (Cd) concentrations due to repeated applications of Cd-rich superphosphate fertiliser (Bramley 1990). The average concentration of Cd in dry topsoil of New Zealand pasturelands is 0.44 mg/kg (Roberts et al. 1994). Robinson et al. (2000) showed that a commonly used willow clone, Tangoio (Salix matsudana x S. alba), accumulated Cd at levels of up to 14 mg/kg in the dry leaves when grown in a soil containing just 0.6 mg/kg of this element. This concentration is above levels (1-5 mg/kg) shown to adversely affect livestock (Underwood and Suttle 1999). Thus Cd accumulation by willows that might be used for fodder is of concern. Willows may also facilitate the entry of Cd into the food chain via insect browsing. The recently introduced willow saw fly (Nematus oligospilus) can completely defoliate willows, going through several generations in 1 year (Charles and Allan 2000). This could introduce another exposure pathway for Cd.

The ability of willows to accumulate Cd has been exploited to remove Cd from contaminated soils (Ostman 1994; Greger and Landberg 1997, 1999; Robinson et al. 2000). The phytoremediation of Cd-contaminated soil would involve short rotation coppicing of willows. Harvested material could then be burnt in an incinerator equipped with a Cottrell precipitator (Graham and Cragg 1959) that prevents the metal being lost in the smoke. The remaining ash product could be stored in an appropriately designed landfill. The energy produced by the incineration of the crop might even be utilised for generating electricity (Nixon et al. 2001).

The efficacy of willows to extract soil-metals is due to a number of attributes including high biomass production, and water-use (Chmelar 1973). Mills et al. (2000) demonstrated that within a single clone, Cd uptake was influenced primarily by tree water-use. It has been demonstrated that there is a wide variation in the ability of different willow clones to accumulate heavy metals in the bark and wood (Riddell-Black et al. 1997). Using hydroponic experiments, Punshon and Dickinson (1999) showed interclonal variation in tolerance to Cu, Cd, Ni, and Zn, as well as in the uptake of Cu.

Given the potential economic importance to New Zealand of Cd accumulation by willows when used as stock fodder or in phytoremediation, an investigation was warranted into Cd accumulation in the leaves and wood of New Zealand bred willow clones. This study aimed to determine variation in Cd, Zn, Mn, and Fe accumulation, as well as water use, in 15 willow clones that have been bred or selected for soil conservation purposes in New Zealand. It was anticipated that suitable clones could be found for both stock fodder (low Cd accumulation), and phytoremediation (high Cd accumulation). Thus clones could be matched for soil conditions, whether that soil was pastureland or a contaminated site.

Materials and methods

Experimental design

The experiment was conducted at the Horticultural and Food Research Institute of New Zealand Ltd (HortResearch), Palmerston North, New Zealand (40.2[degrees]S, 175.4[degrees]E). Fifteen willow clones were selected for the experiment from the HortResearch poplar and willow nursery at Aokautere. Eight were shrub willows (sub-genus Caprisalix) and 7 were tree willows (sub-genus Salix). Information on the clones is shown in Table 1. Plant material and soils were taken from a stool bed containing 9-year-old willows growing in a fine sandy loam (Manawatu series) with pH 5.7, CEC (cation exchange capacity) 13.4 [cmol.sub.c]/kg, organic carbon content of 63 g/kg, and a resident Cd concentration of 0.3 mg/kg. No further Cd was added to the soil. The soil was homogenised using a soil mixer, and Osmocote fertiliser added at rates recommended by the manufacturer.

Three cuttings of each willow clone were grown separately in 20-L plastic buckets from 5 January 2000 to 8 April 2000. Each planted pole was 200-300 mm in length and 20 mm in diameter. Plants were grown in a shade-house in a randomised block design and watered twice daily using an automatic watering system, dispensing c. 1 L of water per application. The first leaves appeared after 8 days, and at the end of the experiment, the average dry weight of the new growth was 61.6 g/plant.


From 23 March 2000 to 3 April 2000, the daily water-use of individual plants was measured. Automatic watering was discontinued and the buckets were weighed daily using a balance accurate to 0.001 kg. The pots were watered manually on alternate days according to each plant's water use. A control pot containing bare soil was used to calculate soil evaporation.

Sample preparation and metal determination

On 4 April 2000, each tree was defoliated, and all current season's wood was removed. Total leaf area per tree was evaluated using a leaf-area meter (LICOR 3000). Leaves and stems of each cutting were placed in a drying cabinet at 80[degrees]C until a constant weight was reached. These samples were then weighed separately and ground. A 0.15-g subsample was placed into a 50-mL Erlenmeyer flask and digested with 10 mL of concentrated nitric acid. These flasks were placed on a heating block until a final volume of 3 mL was reached. It was then diluted to 10 mL in a measuring cylinder using deionised water and stored in polyethylene containers. Cadmium determinations were performed using a graphite furnace atomic absorption spectrometer (GBC 909 AA). Zinc, Fe, and Mn were determined using flame atomic absorption spectroscopy (GBC Avanta [SIGMA]).

Data analysis

Data from the metal concentration measurements were analysed using MINITAB. Fisher's least significant difference method was used to indicate significant difference between the metal concentrations of different clones. The Spearman Rank Correlation test was then used to test rankings of clones given water-use, and Cd, Zn, Mn, and Fe concentration data.

Results and discussions

Cd accumulation

Table 2 summarises the average metal concentrations from all 15 willow clones. Clearly the concentration of Cd is higher in the leaves of the willows than in the stems. There was, however, a significant positive correlation (r = 0.79, P < 0.001) between leaf and stem Cd concentrations. The higher Cd concentrations in the leaves compared with the stems is consistent with the findings of Greger and Landberg (1997) and Robinson et al. (2000). The data also clearly illustrate the ability of some willows to take up significant amounts of Cd from soil, with a mean bioaccumulation coefficient of 13 (Table 2). The bioaccumulation coefficient, as used here, is defined as the leaf/soil metal concentration quotient on a dry weight basis. Given that the Cd concentration in the soils of this experiment (0.3 mg/kg) is lower than the national average (0.44 mg/kg), the Cd concentrations in willow trees used for soil conservation and fodder purposes may well exceed our experimental values.

All the clones tested had Cd concentrations above levels (1-5 mg/kg) shown to adversely affect livestock (Underwood and Suttle 1999). It is unclear, however, what adverse effect (if any) willow leaves may have if they only form a small part of the animal's diet. Not all the Cd from the ingested leaves will be adsorbed by the animal's gut. The nutritional implications of using willow for stock fodder warrants further research.

Figure 1a illustrates the mean leaf Cd concentration for each clone. Corresponding bioaccumulation coefficients ranged from 6 to 33. Shrub willows (sub-genus Caprisalix) had significantly (P < 0.01) higher leaf Cd concentrations than tree willows (sub-genus Salix) with mean values of 5.81 and 3.05 mg/kg, respectively.


Accumulation of Zn, Mn, and Fe

Evaluation of the leaf concentrations of Zn, Mn, and Fe was sought to provide additional information about the potential fodder value of different willow clones, and provide clues as to how and why there are differences in Cd accumulation between clonal types. Significant clonal variation was found for Mn and Zn (Fig. 1b, c) but not for Fe (Fig. 1d). These 3 elements were concentrated mostly in the leaves (Table 2). With the exception of Clone 2, the Zn bioaccumulation coefficient in the leaves was > 1, whereas the Mn and Fe bioaccumulation coefficients were <1. As with Cd, shrub willows showed a higher level of Zn and Mn accumulation than tree willows. Average leaf metal concentrations for shrub and tree willows were, respectively, 391 and 159 mg/kg for Mn, 153 and 141 mg/kg for Zn, and 291 and 236 mg/kg for Fe.

Experiments investigating Cd toxicity revealed direct effects on both Mn and Zn metabolism in calves and sheep and showed that Zn consumption may annul toxic Cd effects (Van Bruwaene et al. 1986). Willows with a high Zn concentration may therefore be selected for fodder use so as to reduce Cd toxicity to stock. In addition, supplemental feeding of willow leaves to stock may alleviate Zn deficiency in areas with a low soil Zn concentration. Clark and Millar (1983) reported a Zn concentration of <5 mg/kg in New Zealand pasture species, compared with the mean Zn concentration in willow leaves of 148 mg/kg found here.

Metal accumulation in relation to water use and biomass production

Figure 1e illustrates plant water use per unit of leaf area. The Spearman Rank Correlation test of data presented in Fig. 1a-d (leaf metal concentrations) and Fig. 1e (water-use per unit leaf area) showed no significant relationships. This finding indicates that differences in metal accumulation between clones are controlled mostly by factor(s) other than water use. Our findings differ from those of Mills et al. (2000) because we investigated the relationship between Cd concentration and water-use over several different clones whereas Mills et al. (2000) investigated one single clone. There was no significant relationship between plant biomass production (Fig. 1f) and either leaf or stem metal concentration.

Caution should be applied when extending our results to the field. The clonal variations with prospects for fodder or phytoremediation are nonetheless exciting and merit further exploration. Our experiments, however, conducted over one growing season, may not represent the long-term metal accumulation behaviour of the clones. The root morphology and soil-Cd distribution in the field need not bear any resemblance to our buckets filled with homogenised soil. The accumulation of metals could be affected by different soil types and the concentrations of other elements (Punshon and Dickinson 1997). The soil used in our experiments has a relatively low clay-fraction and organic matter content. This may have resulted in increased metal bioavailability and plant uptake, relative to other soils. The presence or absence of arbuscular mycorrhizal fungi may also play a role in the metal accumulation and tolerance of Salix (Harris and Jurgensen 1977).


Willow clones have been found to display a large interclonal variation with respect to Cd accumulation from soil. The most efficient clones for Cd-extraction have leaf Cd concentrations 6 times higher than the lowest one and they can concentrate Cd within leaves at levels 33 times that of the soil. These clonal differences cannot be attributed to either biomass or water-use. They may be due to some genetic expression and could be related to differing root microflorae. Shrub willows generally showed greater concentrations of elements within their leaves when compared with tree willow.

Willows with both high-biomass production and low leaf Cd concentrations, such as clones 11 and 13, can be selected for stock fodder on Cd-contaminated soils. These characteristics are typically found in tree willows. Shrub willows, such as clones 1 and 5, showed good biomass production, and their Cd concentration in their leaves tended to be higher indicating their suitability as phytoremediation plants for Cd-contaminated soil. Our results illustrate the potential for manipulating the characteristics of specific willow types to improve the appropriateness and efficacy for particular land management strategies.


Thanks to Vaughan Cruden for technical support and to Alistair Hall for assistance with statistical analysis. Harold van Es provided a thorough review of the manuscript. This work was conducted under the FRST contract C06X004 `Knowledge Tools for Environmental Action'.


Bramley RGV (1990) Cadmium in New Zealand agriculture. New Zealand Journal of Agricultural Research 33, 505-519.

Charles JG, Allan DJ (2000) Development of the willow sawfly, Nematus oligospilus, at different temperatures, and an estimation of voltinism throughout New Zealand. New Zealand Journal of Zoology 27, 197-200.

Chmelar J (1973) Propagation of willows by cuttings. New Zealand Journal of Forestry Science 4, 185-190.

Clark RG, Millar KR (1983) Cobalt. In `The mineral requirements of grazing ruminants.' (Ed. ND Grace) pp. 27-38. (New Zealand Society of Animal Production: Hamilton)

Douglas GB, Bulloch BT, Foote AG (1996) Cutting management of willows (Salix spp.) and leguminous shrubs for forage during summer. New Zealand Journal of Agricultural Research 39, 175-184.

Graham RP, Cragg LH (1959) Suspensions and colloidal dispersions. In `The essentials of chemistry'. (Eds RP Graham, LH Cragg) pp. 490-502. (Clarke, Irwin & Company Limited: Toronto)

Greger M, Landberg T (1997) Use of willow clones with high Cd accumulating properties in phytoremediation of agricultural soils with elevated Cd levels. In `Contaminated soils: 3rd International Conference on the Biogeochemistry of Trace Elements'. Paris, France, 15-19 May, 1995. (Ed. R Prost) pp. 505-511. (Institut National de la Recherche Agronomique (INRA): France)

Greger M, Landberg T (1999) Use of willow in phytoextraction. International Journal of Phytoremediation 1, 115-123.

Harris MM, Jurgensen MF (1977) Development of Salix and Populus mycorrhizae in metallic mine tailings. Plant and Soil 47, 509-517.

Hathaway RL (1986) Short-rotation coppiced willows for sheep fodder in New Zealand. New Zealand Agricultural Science 20, 140-142.

Mills T, Robinson B, Green S, Clothier B, Fung L, Hurst S (2000) Difference in Cd uptake and distribution within poplar and willow species. In `Proceedings of the 42nd Annual Conference and Expo of the New Zealand Water and Waste Association'. Rotorua, New Zealand.

Nixon DJ, Stephens W, Tyrrel SF, Brierley EDR (2001) The potential for short rotation energy forestry on restored landfill caps. Bioresource Technology 77, 237-245.

Ostman G (1994) Cd in Salix--a study of the capacity of Salix to remove Cd from arable soils. In `Willow vegetation filters for municipal wastewaters and sludges: A biological purification system. Proceedings of a study tour, conference and workshop in Sweden'. (Eds P Aronsson, K. Perttu) pp. 153-155. (Sveriges Lantbruksuniversitet: Sweden)

Punshon T, Dickinson N (1997) Acclimation of Salix to metal stress. New Phytologist 137, 303-314.

Punshon T, Dickinson N (1999) Heavy metal resistance and accumulation characteristics in Willows. International Journal of Phytoremediation 1, 361-385.

Riddell-Black D, Pulford ID, Stewart C (1997) Clonal variation in heavy metal uptake by willow. Aspects of Applied Biology 49, 327-334.

Roberts AHC, Longhurst RD, Brown MW (1994) Cadmium status of soils, plants and grazing animals in New Zealand. New Zealand Journal of Agricultural Research 37, 119-129.

Robinson BH, Mills TM, Petit D, Fung LE, Green SR, Clothier BE (2000) Natural and induced Cd-accumulation in poplar and willow: Implications for phytoremediation. Plant and Soil 227, 301-306.

Underwood EJ, Suttle NF (1999) `The mineral nutrition of livestock.' 3rd edn. pp. 531-533. (CAB International: Wallingford, UK)

Van Bruwaene R, Kirchmann R, Impens R (1986) Cd contamination in agricultural and zootechnology. In `Cd in the environment'. (Eds H Mislin, O Ravera) (Birkhauser: Basel, Boston, Stuttgart) Experientia Supplementum 50, 87-96.

Wilkinson AG, Zsuffa L, Verwijst T (1999) Poplars and willows for soil erosion control in New Zealand. Biomass-and-Bioenergy 16, 263-274.

Manuscript received 18 February 2002, accepted 2 July 2002

Thierry Granel (A), Brett Robinson (AB), Tessa Mills (A), Brent Clothier (A), Steve Green (A), and Lindsay Fung (A)

(A) The Horticulture and Food Research Institute of New Zealand, Private Bag 11 030, Palmerston North, New Zealand.

(B) Corresponding author; email:
Table 1. Willow clones used in the experiments

Clone    Cross                                         Tree/shrub   Sex

 1       S. viminalis (L.)                             Shrub        M
 2       S. opaca (Anderss. ex. Seemen)                Shrub        M
 3       S. caprea (L.) x S. viminalis (L.)            Shrub        F
 4       S. viminalis (L.)                             Shrub        F
 5       S. schwerinii (E. Wolf)                       Shrub        M
 6       S. disperma                                   Shrub        F
 7       S. dasyclados (Wimm.)                         Shrub        F
 8       S. triandra (L.)                              Shrub        M
 9       S. matsudana (Koidz) x S. alba (L.) clone 1   Tree         F
10       S. matsudana (Koidz) x S. alba (L.) clone 2   Tree         F
11       S. matsudana (Koidz) x S. alba (L.) clone 3   Tree         M
12       S. matsudana (Koidz) x S. alba (L.) clone 4   Tree         F
13       S. babylonica (L.) x S. alba (L.) clone 1     Tree         F
14       S. babylonica (L.) x S.alba (L.) clone 2      Tree         F
15       S. fragilis (L.)                              Tree         F

Table 2. Metal concentrations (mg/kg dry matter) in leaves, stems, and
associated soils of the willow clones

                                Cd      Mn       Zn        Fe

Leaf                 Median     3.9    236      137      136
                     s.e.       0.5     32        8       70
Stem                 Median     2.7     65       54       20
                     s.e.       0.2     10        4        4
Soil                            0.3    597       64        5.6%
  coefficient (A)              13.0      0.4      2.1      0.002

(A) The leaf/soil metal concentration quotient on a dry weight basis.
Gale Copyright: Copyright 2002 Gale, Cengage Learning. All rights reserved.