Root development and heavy metal phytoextraction efficiency: comparison of different plant species in the field

Heavy metal phytoextraction is a soil remediation technique which implies the optimal use of plants to remove contamination from soil. Plants must thus be tolerant to heavy metals, adapted to soil and climate characteristics and able to take up large amounts of heavy metals. Their roots must also fit the spatial distribution of pollution. Their different root systems allow plants to adapt to their environment and be more or less efficient in element uptake. To assess the impact of the root system on phytoextraction efficiency in the field, we have studied the uptake and root systems (root length and root size) of various high biomass plants (Brassica juncea, Nicotiana tabacum, Zea mays and Salix viminalis) and one hyperaccumulator (Thlaspi caerulescens) grown in a Zn, Cu and Cd contaminated soil and compared them with total heavy metal distribution in the soil. Changes from year to year have been studied for an annual (Zea mays) and a perennial plant (Salix viminalis) to assess the impact of the climate on root systems and the evolution of efficiency with time and growth. In spite of a small biomass, T. caerulescens was the most efficient plant for Cd and Zn removal because of very high concentrations in the shoots. The second most efficient were plants combining high metal concentrations and high biomass (willows for Cd and Zn and tobacco for Cu and Cd). A large cumulative root density/aboveground biomass ratio (L_A/B), together with a relative larger proportion of fine roots compared to other plants seemed to be additional favourable characteristics for increased heavy metal uptake by T. caerulescens. In general, for all plants correlations were found between L _A/B and heavy metal concentrations in shoots (r=0.758^***, r=0.594^***, r=0.798^*** (P<0.001) for Cd, Cu and Zn concentrations resp.). Differences between years were significant because of variations in climatic conditions for annual plants or because of growth for perennial plants. The plants exhibited also different root distributions along the soil profile: T. caerulescens had a shallow root system and was thus best suited for shallow contamination (0.2 m) whereas maize and willows were the most efficient in colonising the soil at depth and thus more applicable for deep contamination (0.7 m). In the field situation, no plant was able to fit the contamination properly due to heterogeneity in soil contamination. This points out to the importance and the difficulty of choosing plant species according to depth and heterogeneity of localisation of the pollution.