*2.5. Metals Determination*

For the determination of the metals under study in soils, 2 g of each sample were subjected to an acidic digestion process as described in ISO 11466:2010 [31]. The extraction of the metal content was conducted with *aqua-regia*, using a conventional heating process (open sand bath system) at (120 ± 10) ◦C. The residue from the acidic digestion was filtrated using a cellulose filter and diluted to a final volume of 100 mL with ultrapure water (18.2 M Ω cm).

Each metal content in the soil samples was determined by GFAAS, in a Thermo Elemental SOLAAR, Suite M5 Spectrometer, equipped with an automatic sampler FS95 and a graphite furnace GF95, controlled by SOLAAR software, v.11.2.

#### *2.6. Evaluation of Environmental Risks—Pollution and Ecological Index Models*

Determining the concentration of toxic metals in soil is not enough to assess the state of soil pollution. It is common to use several indexes to estimate the degree of soil pollution and the ecological risk of the soil for humans and biota. The most important are the enrichment and contamination factors, the pollution and geo accumulation indexes and the ecological risk factor. These indexes are presented and explained below according to their original definition in the references cited for each one.

## 2.6.1. Geo-Accumulation Index (*I*geo)

The Geo-accumulation index (*I*geo) is calculated with the formula proposed by Förstner and Müller [32]. This relationship (Equation (1)) is used to assess the degree of contamination of soil and water by metals.

$$I\_{\rm geo} = \log\_2 \frac{\mathcal{C}\_{\rm M}}{1.5 \times \mathcal{C}\_{\rm B}} \tag{1}$$

In Equation (1) *C* M represents the concentration of the metal in the soil and *C*B represents the background concentration of that same metal in unpolluted soils.

According to Förstner and Müller [32], *<sup>I</sup>*geo values are categorized into seven contamination classes ranging from Class 0 (*I*geo < 0) classified as "uncontaminated" until Class 6 (*I*geo > 5) classified as "extremely contaminated".

It should be noted that *<sup>I</sup>*geo value greater than 6 is an indication of soils with 100 times the metal concentration considered as background value, a very high pollution by the metal under study.

#### 2.6.2. Contamination Factor ( *C*f)

The contamination factor defined almost simultaneously by Tomlinson et al. [22] and Hakanson [23] was used to measure levels of metal pollution in sediments in lakes. However, these factors were widely adopted for the calculation of pollution indexes also in soil. The factor is defined as the ratio between the metal concentration in the soil ( *C* M) and the reference concentration of that same metal in unpolluted soil ( *C*b), the background values or preindustrial reference values (Equation (2)).

$$C\_{\rm f} = \frac{C\_{\rm M}}{C\_{\rm b}} \tag{2}$$

Considering the calculated index, soils can be classified according to the contamination levels proposed by Hakanson [23] from "low contamination" ( *C*f < 1) to "very high contamination" ( *C*f ≥ 6).

#### 2.6.3. Degree of Contamination ( *C*d)

The degree of contamination ( *C*d), also defined by Hakanson [23], represents the sum of the contamination factors. While the contamination factor represents the individual contribution of each metal, the degree of contamination considers all the *n* polluting metals in a given location, and is defined by Equation (3):

$$\mathbb{C}\_{\mathbf{d}} = \sum \mathbb{C}\_{\mathbf{f}} \tag{3}$$

The classification used for this factor was also proposed by Hakanson [23] and goes from "low degree of contamination" for *C*d < 8 to "very high degree of contamination" for *C*d ≥ 32 indicating serious anthropogenic pollution.

#### 2.6.4. Pollution Load Index (*PLI*)

The Pollution load index (*PLI*) proposed by Tomlinson et al. [22] measures the overall degree of contamination in the sample and is often used to calculate the degree of soils pollution and other polluted ecosystems. For a site or zone, it is calculated using Equation (4):

$$PLI\_{\text{zone}} = \sqrt[n]{\mathbb{C}\_{\text{f}}(1) \times \mathbb{C}\_{\text{f}}(2) \times \mathbb{C}\_{\text{f}}(3) \times \dots \times \mathbb{C}\_{\text{f}}(n)}\tag{4}$$

Adapting the concept defined for the calculation of PLI for an estuary, the PLI for a city can be obtained using the PLI of each zone (Equation (5)).

$$PLI\_{\text{city}} = \sqrt[n]{PLI\_{\text{zone 1}} \times PLI\_{\text{zone 2}} \times PLI\_{\text{zone 3}} \times \dots \times PLI\_{\text{zone n}}} \tag{5}$$

2.6.5. Ecological Risk Factor (*E*r) and Global Potential Ecological Risk (*RI*)

The ecological risk factor (*E*r) proposed by Hakanson [23] is used to assess the ecological risk of an element in the soil and can be obtained using Equation (6):

$$E\_\mathbf{r}^1 = T\_\mathbf{r}^1 \times \mathbb{C}\_\mathbf{f}^1 \tag{6}$$

where *C*i f is the contamination factor of element i and *T*i r is the toxic response factor of element i. The toxic response factors for Cd, Cr, Ni and Pb are 30, 2, 5, 5, respectively [23].

*E*r values proposed by Hakanson [23] are divided in five ranges from "low potential ecological risk" for *E*r < 40 to "very high ecological risk" when *E*r > 320. The sum of the individual potential risks (*E*<sup>i</sup> r) gives the global potential ecological risk (*RI*) for the soil (Equation (7)).

$$RI = \sum E\_\mathbf{r}^\mathbf{i} = \sum T\_\mathbf{r}^\mathbf{i} \times C\_\mathbf{f}^\mathbf{i} \tag{7}$$

Originally, the global ecological risk was a diagnostic tool for water pollution control but presently it has been used with success for assessing quality of sediments and soils contaminated by heavy metals [33–36]. The classification of the global ecological risk (*RI*)*,* according to Hakanson [23], varies from "low global ecological risk" (*RI* < 150) to "very high global ecological risk" (*RI* > 600).

#### *2.7. Quality Control and Quality Assurance*

Analyses were performed carrying out quality assurance and quality control procedures to provide high analytical precision. Special care was taken to minimize crosscontamination, contamination of glass material. Additionally, all efforts were made to minimize contamination from air.

The GFAAS operating conditions were optimized for each metal using reference metal standard solutions, CertiPUR® from Merck (Darmstadt, Germany). Sample concentrations were determined using calibration curves evaluated in terms of linearity and sensitivity, according to the obtained correlation coefficient and slope, respectively. Limits of detection and quantification for each metal were determined. The validity of the calibration curves was performed daily using control standards prepared using CertiPUR® standards from Merck in the same acid matrix used for soil samples. For the control standards the same procedure analysis used for real samples was applied. Two control standards were used to test the precision of the calibration curve of each metal in two points: ~25% and 75% of the linear range. The precision of the calibration curve was maintained between 2% and 5% and never higher than 8%.

The extraction method was validated through a soil certified reference material (BCR CRM-142R Light Sandy Soil) from IRMM (Institute for Reference Materials and Measurements). The reference soil was subjected to the same acid digestion procedure as the soil samples and the recovery percentage values of the CRM were 93% for Pb, 106% for Cd, 93% for Ni and 0% for Cr (since the CRM did not contain chromium).

All samples were digested and analyzed in duplicate, and the final content was always considered as the average value of six measurements made in repeatability conditions.
