Chemosphere 119 (2015) 217–223
Contents lists available at ScienceDirect
Chemosphere journal homepage: www.elsevier.com/locate/chemosphere
Screening biological traits and fluoride contents of native vegetations in arid environments to select efficiently fluoride-tolerant native plant species for in-situ phytoremediation Asma Boukhris a,b, Isabelle Laffont-Schwob a,⇑, Imed Mezghani b, Lefi El Kadri b, Pascale Prudent c, Anca Pricop a,c, Thierry Tatoni a, Mohamed Chaieb b a Aix Marseille Université, Avignon Université, CNRS, IRD, Institut Méditerranéen de Biodiversité et d’Ecologie marine et continentale (IMBE), 52 avenue Normandie-Niemen, 13397 Marseille cedex 20, France b Plant diversity and Ecosystems in Dry Environment, Faculty of Science, University of Sfax, 3000 Sfax, Tunisia c Aix Marseille Université, CNRS, LCE FRE 3416, Laboratoire de Chimie de l’Environnement, case 29, 3 place Victor Hugo, 13331 Marseille cedex 3, France
h i g h l i g h t s Emphasing a gradient of fluoride pollution in soils of the arid East-coast of Tunisia. Statement of a fluoride transfer to the aerial parts of steppic native vegetation. Specific biological traits of the native plant species are useful for plant selection. Proposal of an efficient approach for plant screening for in-situ phytoremediation.
a r t i c l e
i n f o
Article history: Received 26 February 2014 Received in revised form 19 May 2014 Accepted 2 June 2014
Handling Editor: X. Cao Keywords: Fluoride pollution Pollutant tolerance Functional traits Phytostabilization
a b s t r a c t High fluoride pollution has been detected in the surrounding soils of the coastal superphosphate industries in the Gulf of Gabes (Southeast of Tunisia). A study was conducted in vicinity of factories analysing plant functional traits combined with plant fluoride accumulation and soil metal concentrations aiming to screen more efficiently native plant species tolerant to this pollution. Aerial parts of 18 plant species out of the 10 most abundant species per site were harvested on two polluted sites of Gabes and Skhira at the vicinity of the factories and on the less polluted site of Smara. Native plant species accumulated fluoride following the gradient of soil pollution. Fluoride contents of plant aerial parts ranged from 37 mg kg1 to 360 mg kg1 and five plant species were only found in the most polluted site. However these latter had low biomass and soil cover. Crossing biological traits and fluoride contents, a selection grid for potentially restorative plant species enabled the selection of three native perennials i.e. Rhanterium suaveolens, Atractylis serratuloides and, Erodium glaucophyllum as potential candidates for an in-situ phytoremediation program on arid fluoride-polluted sites. This approach may be used in other fluoride-polluted Mediterranean environments. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Under arid conditions, plant perennials play a key role in the functioning and services of the ecosystem reducing soil erosion. Along the East-coast of Tunisia, the steppe shrub vegetation with biological traits adapted to drought (Jauffret and Visser, 2003), is submitted to stressful and disturbing environment. Since the fifties, the chemical industry of phosphogypsum has developed in some Tunisian cities along this coast. A main source of ⇑ Corresponding author. Tel.: +33 4 13 55 12 30. E-mail address:
[email protected] (I. Laffont-Schwob). http://dx.doi.org/10.1016/j.chemosphere.2014.06.007 0045-6535/Ó 2014 Elsevier Ltd. All rights reserved.
environmental pollution is due to industries of phosphates rock mining and processing, with release of trace metals and fluoride in particular. Rain and storage water can seep through the stockpiled phosphogypsum, carrying the contaminants into the soil and groundwater; air pollution also occurs due to wind-dispersed waste products. More precisely, in the manufacturing process and during the phosphate attack by sulphuric and phosphoric acids, fluoride compounds such as HF, H2SiF6, and CaF2 release off by the factory chimneys (Rouis and Ballivy, 1990; Tayibi et al., 2009). In addition, analysis of the air surrounding the factory of Sfax has shown that fluoride air contents ranged between 3 and 12 lg dm3 d1 (Mezghani et al., 2005; Ben Abdallah et al.,
218
A. Boukhris et al. / Chemosphere 119 (2015) 217–223
2006a). Fluoride pollution is always concomitant with heavy metal contamination (Rouis and Ballivy, 1990; Tayibi et al., 2009), causing a mixed pollution of soils. Phytoremediation is an emerging approach for the remediation of polluted sites, using plants and their associated micro-organisms (Pilon-Smits, 2005). In the context of the fluoride pollution of the East coast of Tunisia, two avenues are worth exploring: one using plants as a vegetation barrier aiming at mitigating polluted particle transfer to the surroundings, and another using plants for pollutant containment in soils namely phytostabilization. For both, targeting potential native plant species to limit the dispersion of this phytotoxic pollutant (Fornasiero, 2001), and avoiding introduction of potentially invasive plant species that may result in decreasing regional plant diversity (Mendez and Maier, 2008), seem to be an innovative idea for in situ polluted sites’ phytoremediation. However, as reported by Roccotiello et al. (2014), only a relatively small number of Mediterranean plant species is being studied for phytoremediation purposes. Moreover, although fluoride effects on crops such as vineyards (Ben Abdallah et al., 2006b) and, pomegranate and almond orchards (Ben Abdallah et al., 2006a), have already been evaluated on some of these sites, there is a lack of knowledge about the impact of this pollution on natural vegetation in the vicinity of the factories, and a fortiori, on their fluoride tolerance. The sites of Gabes and Skhira were selected and compared to Smara, the reference site. On all three sites, the functional traits of vegetation differed related to environmental features, and analysis of life history data showed the reduction of plant cover of perennials to only 5%. A continuous process of degradation of these steppes has been demonstrated by Hanafi and Jauffret (2008). However, this environmental deterioration has never been related to industrial pollution. Thus, there is no information on potential tolerance of the native species to this pollution. Therefore, as recently reported by Schellberg and Pontes (2012) and Pérez-Harguindeguy et al. (2013), gathering species functional traits is a recent promising approach to better understand the ecological functioning of ecosystems. Consequently, in this study, we combined the level of fluoride accumulation in native plant species with information of their life form, dispersal and dissemination, as well as their palatability and ecological preferences, aiming at selecting the most accurate local plants species to support a phytoremediation process on fluoride-polluted soils in arid environments. 2. Materials and methods 2.1. Study area The three studied sites are located along the East coast of Tunisia (Fig. 1). The sites of Gabes and Skhira correspond to a long plain along the coast (30 m above sea level for both sites) and are nearby industrial phosphate factories. The site of Smara (15 m above sea level) is 70 km and 20 km from Gabes and Skhira, respectively and, in similar geochemical, ecological and climatic conditions as both former sites. All the sites are submitted to an arid climate with low average rainfall (from 167 to 176 mm average annual pluviometry with average annual temperature from 18.8 to 19.3 °C). Their distance to the sea were 1.8, 2.8 and 4.3 km for Skhirra, Gabes and Smara, respectively, and they are thus not submitted to seasprays. Climatic parameters were derived from the Wordclim database (Hijmans et al., 2005). 2.2. Floristic survey First, we compared the spontaneous floristic structure and composition of Gabes and Skhira with those in Smara in March 2011, on one ha surface area per site. The vegetation in the three sites
is typical of the arid chamaephytic steppes with Rantherium suaveolens (Le Houérou, 1969), however with different facies depending on the local disturbances and variability of the chemical composition of soils. Therefore, for each site, the ten plant species offering the most important plant cover were selected and at the end of the survey, the lists of these most abundant plant species were compared between the three sites to define which were in common. The plant identification was confirmed at the laboratory using the Flora of arid and Saharan zones (Chaieb and Boukhris, 1998) and voucher specimens are kept at the herbarium of the University of Sfax. 2.3. Plant and soil collections In March 2011, three composite samples of 1 kg of topsoil (0–20 cm depth) were collected at the vicinity of the plant species on each site, to analyze concentration of major and trace elements, including fluoride. Each soil sample was composed of 6 subsamples of equal weight, separated by a distance of 1 m according to a cross pattern on each site. Since the samples were composite samples constituted by equal quantity of 6 mixed subsamples, they could be considered representative of each site. Soil samples were stored in plastic bags until return to the laboratory, where they were dried at room temperature until analysis. Three individuals of the 10 most abundant plant species were sampled by a random harvest i.e. three replicates of 30 samples out of 18 plant species. Only leaves and stems of the middle of the shoots were taken and stored in paper bags until return to the laboratory. Plant samples were then dried at 80 °C during 72 h in an oven (Memmert). 2.4. Fluoride analysis Each composite sample of soil and each plant sample was analyzed in triplicates. Soil samples were sieved to 2 mm mesh. Five grams of soil subsamples were mixed with HCl:H2O (1:1) during 90 min. Then, HCl extract was mixed with total ionic strength-adjustment buffer and analyzed using a fluoride-specific ion electrode (inlab/Model WTW) coupled to a pH-meter (pH ION R503) (Mezghani et al., 2005). Dried plant samples were ground (