Factors determining Zn availability and uptake by plants in soils developed under Mediterranean climate

Highlights

• DTPA extractable Zn had little value to predict Zn deficiency.

• Clay and Olsen P negatively affected Zn uptake in calcareous soils.

• In non-calcareous soils, crystalline Fe oxides negatively affected Zn uptake.

• Estimate of Zn uptake requires discriminating soils according to its carbonate content.

• Soil properties related to Zn adsorption should be considered in Zn uptake estimation.

Abstract

Zinc deficiency is an extended agronomic problem, particularly in staple food crops such as cereals. The availability of Zn to plants is ruled by soil properties, biological factors in the rhizosphere, and interaction with other nutrients. These factors may constrain the predictive value of Zn availability indices. This work aimed at assessing the soil factors that affect the absorption of Zn by plants and improving the predictive value of conventional indices. To this end, an experiment was performed using durum wheat (Triticum durum L.) grown on a set of soils developed under Mediterranean climate.

In calcareous soils, Zn uptake by plants decreased with increased clay content and Olsen P (P_Olsen), meanwhile in non-calcareous soils it decreased with increased crystalline Fe oxides content. Biological factors such as microbial activity and organic anion exudation in the rhizosphere contribute to Zn uptake by plants. No relationship was found between Zn uptake by plants and the DTPA extractable Zn (Zn_DTPA). Pyrophosphate extractable Zn (Zn_Pyro) was only related to Zn uptake by plants in calcareous soils (R² = 0.29; P < 0.01). The best estimation of Zn uptake by plants in calcareous soils was obtained with a model involving Zn_Pyro/P_Olsen ratio and clay content (R² = 0.57; P < 0.001). In non-calcareous soils, Zn uptake by plants was accurately estimated with a model involving Fe bound to crystalline oxides and rhizospheric oxalate (81% of the variance explained). Results reveal the need of discriminating soils according to its carbonate content and the use of soil properties related to Zn adsorption capacity, such as clay and Fe oxide content, and Olsen P for accurate estimation of Zn uptake by plants.

Introduction

The deficiency of micronutrients is an extended nutritional imbalance constraining crop production in many agricultural lands in the world (Ryan et al., 2013). Around one-third of cultivated soils have low Zn availability levels for suitable crop production (Broadley et al., 2007, Alloway, 2009, Cakmak and Kutman, 2018). In the Mediterranean basin, Zn deficiency is an extended agronomic problem mostly ascribed to soil chemical properties (e.g. carbonates, pH, low organic matter content) (Rashid and Ryan, 2004, Alloway, 2009). Zn deficiency causes yield decreases, particularly in cereals, which are staple food in many regions of the world (Ryan et al., 2013), frequently without clear deficiency symptoms. Zn deficiency not only implies a productivity constraint, but also a reduction in the grain quality for human consumption, with clear health implications in regions with cereal-based diets (Yang et al., 2007, Cakmak, 2008, Borrill et al., 2014).

The mobility and availability of Zn to plants is ruled by soil properties and biological factors in the rhizosphere (Moreno-Lora et al., 2019). In this regard, the root and microbial exudates may increase the solubility and mobilization of micronutrients as a result of soil acidification or complexing process (Marschner, 1993, Rengel, 2015). Plants absorb Zn from soil solution as free ions, that may came from different pools: soil solution, soluble organic complexes, and that in equilibrium with the soil solution (easily-desorbable) adsorbed on different minerals (Adhikari and Rattan, 2007, Alloway, 2009, Mousavi, 2011). Nevertheless, the available fraction represents only a small portion of total Zn content in soils, while 80–90% is present in the relatively inactive clay lattice and insoluble precipitated forms (Adhikari and Rattan, 2007, Regmi et al., 2010). Zn concentration in soil solution is very low and its mobility and transport to the root surface is limited by the forms in equilibrium with soil solution (Rengel, 2015).

The supply of Zn from soil to plants is assumed to be governed mainly by its concentration in the parental material, soil chemical properties (pH, organic matter, clay minerals, sesquioxides, carbonates), and nutrient interactions such as that with other micronutrients and the well-known antagonism with P (Adhikari and Rattan, 2007, Alloway, 2009, Fageria et al., 2002, Imtiaz et al., 2006, Mousavi, 2011). Sandy soils poor in organic matter are usually poor in total and available Zn, with a limited capacity of replenishing the nutrient absorbed from soil solution (Alloway, 2009, Rashid and Ryan, 2004, Ryan et al., 2013). In calcareous soils, co-precipitation on calcite and chemisorption on Fe oxides are relevant process explaining Zn fixation (Uygur and Rimmer, 2000, Montilla et al., 2003, Buekers et al., 2007, Alloway, 2009, Rengel, 2015, Ryan et al., 2013). Free CaCO₃ not only contributes to fix the nutrient by specific and non-specific adsorption onto calcite (Montilla et al., 2003), but also the ensuing high pH reduces its solubility by enhancing the precipitation of Zn hydroxides (Rashid and Ryan, 2004), and by promoting an essentially non-reversible adsorption on Fe oxides (Buekers et al., 2007, Buekers et al., 2008).

Fe oxides adsorb Zn through surface complexation in an irreversible process at pH > 6, and by non-specific adsorption which is dependent on pH since negative surface charges increases at increased pH (Buekers et al., 2007, Buekers et al., 2008, Stahl and James, 2010, Ryan et al., 2013). The effect of an increased specific surface in Fe oxides, higher in poorly crystalline than in crystalline ones, is a major factor explaining the role of the former in Zn adsorption by surface complexation (Komárek et al. 2018), which increases with reaction time (Shuman, 1977, Buekers et al., 2008, Duffner et al., 2014). In other minerals, such as kaolinite, surface complexation is also deemed to be the dominant mechanism for Zn adsorption (Wang et al. 2017).

It is well known that high P availability levels in soils may negatively affect Zn uptake by plants (Loneragan et al., 1979, Zhang et al., 2012). However, this antagonistic effect seems to vary depending on the different soil minerals that adsorb P and Zn (Sánchez-Rodríguez et al., 2017). Thus, the risk of Zn deficiency induced by P seems to be increased at increased Fe oxides to CaCO₃ ratio (Rahmatullah, 2000). Phosphate competition for the adsorption sites may induce Zn deficiency in soils (Ryan et al., 2013). This competition however, is affected by the availability of sorbent sites, which depends on the content of Fe oxides and their crystallinity.

Despite the body of literature published, many aspects of Zn dynamics in soils are not fully understood. Given the relevance of carbonates and Fe oxides in soils, better understanding of their role in Zn reactions and availability is necessary for predicting and managing Zn deficiency in agricultural soils. Despite evidences on the contribution oxides and carbonates to Zn adsorption in soils (Ji and Cang, 1993, Montilla et al., 2003), the role of crystalline and poorly crystalline oxides in basic soils, or the relative contribution of oxides and carbonates to Zn availability remains to some extent not understood. This role of adsorbent surfaces is affected by the interaction between Zn and P since these nutrients are adsorbed on the same minerals, and even may compete for adsorption sites (Zhao and Selim, 2010). On the other hand, the coadsorption of P and Zn has proved crucial explaining Zn retention on poorly crystalline oxides (Liu et al. 2016). Thus, assessment of Zn dynamics should also consider the interaction with P dynamics in soil.

Conventional Zn availability indices for assessing the potential crop response to fertilization are usually based on the use of complexing agents, such as DTPA (Lindsay and Norvell, 1978). However, these methods have sometimes little predictive value (Feng et al., 2005, Maqueda et al., 2015). Interaction with other nutrients and soil properties affecting Zn dynamics may constraint the predictive value of Zn availability indices. As mentioned above, these factors governing the dynamics of Zn, and consequently its availability to plants, are not fully understood. Consequently, for improving the estimates of Zn availability and the prediction of the potential response to Zn fertilizers, further knowledge of soil factors ruling Zn availability to plants is required. It may be hypothesized that availability indices may be improved or corrected in order to increase its predictive value if soil factors affecting Zn availability to plants are taken into account.

On these grounds, the objectives defined for the present work were: (i) to study soil factors affecting Zn uptake by plants in a set of representative soils developed under Mediterranean climate, and (ii) to achieve accurate estimates of Zn uptake by plants by using usual Zn extraction procedures in combination with soil properties.