The experimental study ( Figure 1 ) was carried out at the ALSIA Metapontum Agrobios Research Center, province of Matera (N 40 23´, E 16 47´), Italy. The climate in the region is Mediterranean, with moderate, wet winters, and hot, dry summers. The average temperature in the summer months (June to September) ranges from 24°C to 28°C, while the average temperature in the winter months (December to February) ranges from 5°C to 11°C. The total annual rainfall averages around 600-700 millimeters, with most of the precipitation occurring from November through April.

On 17/06/2021, the tomato cultivar ‘Taylor F1’ (Solanum lycopersicum L.; formerly Lycopersicon esculentum Mill.) was transplanted in weighing lysimeters ( Figure 2 ).

Pre-cultivated tomato seedlings in 180-hole polystyrene honeycomb containers were transplanted into 0.8 m³ tanks at the 3^rd-4^th true leaf stage for each experimental treatment distributed according to the randomized block experimental scheme with four (4) repetitions ( Figure 3 ). The experimental design entailed comparing three irrigation treatments:

TWW effluent from a standard municipal wastewater treatment plant (WWTP) at the experimental site (Ferrandina, Italy) was utilized to determine TWWx1 and TWWx3 irrigation treatments. Rapid sand filtration (rSF) and UV treatment are used for tertiary treatment and disinfection. The experimental design includes four lysimetric measures (plots) for each irrigation treatment.

Among the ECs were clarithromycin, sulfamethoxazole, trimethoprim, carbamazepine, diclofenac, fluconazole, climbazole, ketoprofene, metoprolol, and naproxen. These substances were specifically chosen due to their prevalence in wastewater; they are often not completely eradicated during standard treatments. Table 1 lists the chemical structures and attributes of the selected ECs. The concentration of these EC in treated wastewater ranged from low ng L⁻¹ to low µg L⁻¹ (Ben Mordechay et al., 2021). Standards (> 98% of purity) were used to prepare the multi-compound stock standard solution (1000 ppm). This solution was added to wastewater used for irrigation to achieve the concentration of 200 and 600 μg L⁻¹ of each compound and obtain TWWx1 and TWWx3.

The lysimetric tanks were filled with sandy loams soil (United States Department of Agriculture classification) with the following physical and chemical properties: sand, 84.7%; silt, 3.3%; clay, 12.0%; field capacity (measured by pressure plate apparatus at -0.03 MPa) of 13.2% dry weight (dw); wilting point (measured by pressure plate apparatus at -1.5 MPa) of 7.2% dw, and a bulk density of 1.45 Mg m^-3; pH 8.3; electrical conductivity, 0.10 dS m⁻¹; organic matter, 0.32% (Walkley and Black method); available phosphorus (Olsen method), 35.6 mg kg^-1; total potassium, 0.92 g kg^-1 (determined by coupled plasma optical emission spectrometer, Agilent, ICP-OES 720); total nitrogen, 0.51% (Kjeldahl method); mineral NO₃–N, 0,7 mg kg⁻¹; mineral NH₄-N, 2.7 mg kg⁻¹. This type of soil is characteristic of the Ionian-Metapontine coastline and is extensively employed for vegetable cultivation (Candido et al., 2013). Additionally, this soil has allowed us to operate under favorable hydraulic conductivity conditions, enabling the monitoring of the solution’s movement circulating in the soil through the use of moisture sensors. Three plants were transplanted into each tank, and throughout the cultivation cycle, typical agronomic practices for growing and processing tomatoes in Basilicata were followed. Each lysimeter was periodically irrigated using a micro-flow irrigation system, with drippers installed at each plant, during the cultivation cycle. Following the initial irrigation, which was carried out by applying a volume of water sufficient to return the entire volume of soil to the Field Water Capacity (FWC), a weekly irrigation rotation with an irrigation volume suitable for returning the soil moisture to the FWC was carried out (Allen et al., 1998). The crop water consumption between irrigations was calculated by weighing the individual lysimeter tanks with a trans pallet equipped with load cells. The difference in tank weight between the end of the previous irrigation and the start of the next one represents the water consumption during that time interval as well as the irrigation volume required to restore the soil’s FWC. A probe was inserted in each experimental plot’s lysimeter to test the validity of the irrigation scheduling criterion and maybe correct the specific volume of watering. A scanner outfitted with Diviner 2000 sensors from Sentek Technologies was used to monitor soil moisture. We were able to accurately monitor all components of the water balance and collect drainage water samples to trace any EC movement in the aquifer thanks to the lysimeters. In this regard, because the tomato test takes place in a protected setting, irrigation volume was purposely raised at a given moment during the growing cycle to induce drainage.

The concentration of ECs in water samples (spiked wastewaters and leached waters) was evaluated using an online solid phase extraction (SPE) method using previously established analytical settings (UPLC-QTOF/MS/MS) (Montagna et al., 2020). To extract ECs from soils, the modified QuEChERS method (De Mastro et al., 2022b) was used. Before extracting ECs from various parts of the plant, roots were gently hand washed with tap water to remove soil residues, then rinsed with deionized water and blotted dry with a paper towel. Finely chopped roots, leaves, stems, and tomatoes were stored in a 50-mL centrifuge tube in the dark at -20°C until extraction. In a 50 mL plastic centrifuge tube, 2 g of roots, leaves, and stems or 10 g of tomato fruits were placed and spiked with the appropriate recovery surrogate. Except for the tomatoes, 6 mL of water was added to the centrifuge tubes before capping and vortexing for 1 minute. After thoroughly wetting the samples, 10 mL of Acetonitrile was added to the centrifuge tubes and shaken by hand for 5 minutes. After this step, only the leaves, stems, and fruits were allowed to rest for 15 minutes. After that, a salting out step with Citrate buffer (4 g MgSO₄, 1 g NaCl, 0.5 g NaCitrate dibasic sesquihydrate, 1 g NaCitrate tribasic dihydrate) was performed. For 5 minutes, the tubes were vigorously shaken by hand. Following that, the samples were centrifuged for 5 minutes at 3700 rpm, resulting in a phase separation of the aqueous and organic solvents. The upper ACN layer (6 mL) was transferred into 15 mL tubes for the clean-up step. Tubes containing 900 mg MgSO_4 + 150 mg primary secondary amine (PSA) for roots, 900 mg MgSO_4 + 150 mg PSA + 150 mg octadecyl (C18) for leaves and stems, 900 mg MgSO150 mg PSA + 15 mg graphitized carbon black (GCB) for fruits, were vortexed for 1 min. After centrifugation (5 min, 4000 rpm), the supernatant was filtered through a membrane filter (PVDF, 0.22 μm), and 1.5 mL was transferred into a screw cap vial for LC-MS/MS analysis to determine the concentration of ECs from the four replicates of each thesis.

The ANOVA procedure was applied to all datasets using a randomized complete design with four replicates. A one-way ANOVA procedure (Christensen, 2020) was used with the irrigation typology (FW, TWWx1, and TWWx3) as fixed factors and the replication as random. The entire dataset was tested using the analysis of variance (ANOVA) assumptions. The normality distribution of the model’s residuals was verified graphically (QQ-plot) and statistically (Shapiro-Wilk normality test). Furthermore, Levene’s test was used to confirm homoscedasticity. The experimental design and random sampling for the different matrices met the final ANOVA assumption. When all three ANOVA assumptions were met, the ANOVA was applied to the model. Only when the ANOVA revealed a significant difference (p-value 0.05), was a post hoc analysis of the estimated marginal averages performed using Tukey’s HSD (honestly significant difference) test from the R package agricolae (de Mendiburu and Yaseen, 2020).

Table 2 depicts the main components of water balance (seasonal irrigation volume, rainfall, and drainage), as well as the total ECs intake in the lysimeters. The total amount of applied irrigation water (I.V.) was 620.8 mm, while the total amount of drained water (D) was 25 mm. The total rainfall for the tomato growth cycle (R)was 88 mm. Figure 4 depicts the total amount of water and ECs applied to the soil using the TWWx1 and TWWx3 irrigation treatments.

Figures 5 and 6 depict the final concentrations of ECs in soil and plant matrices (root, stem, leaf, and fruits) at the end of the cultivation cycle. The FW irrigation approach contained no significant concentrations of target ECs. The TWWx1 method acted differently for each matrix ( Figure 5 ). Rather than the fruit, the leaves had high levels of two ECs, fluconazole, and carbamazepine. The residual pollutant amounts in plant tissues were not substantially different from zero. Fluconazole, carbamazepine, and metoprolol levels in plant leaves, roots, stems, and fruits increased significantly with the TWWx3 strategy ( Figure 6 ). The concentrations of the remaining contaminants in plant tissues were not significantly different from zero. The largest quantities of the three pollutants observed in the plants (fluconazole, carbamazepine, and metoprolol) were found in the leaves in both irrigation treatments (TWWx1 and TWWx3), with lower but substantial concentrations reported in the stems, roots, and fruits. Martínez-Piernas et al., 2019 observed similar results, where organic microcontaminant concentrations were lower in tomato fruits, generally 10 times lower in fruit compared to leaves. Significant quantities of climbazole, fluconazole, carbamazepine, sulfamethoxazole, and clarithromycin were discovered in soil irrigated with TWWx1 and TWWx3 water. The other five pollutants in the soil had statistically negligible concentrations ( Figures 5 , 6 ). Results of Pico et al. (2019) study revealed the potential uptake and accumulation by crops of carbamazepine (as 10,11-carbazepine epoxide), atenlolol, caffeine, gemfibrozil and ibuprofen (as ibuprofen hexoside). Some pharmaceuticals and seven pesticides were detected in plants. Pharmaceuticals and ECs were found in quantifiable levels in all irrigation water, soils, and plants (>99.6%) in Israel (Ben Mordechay et al., 2022). Martínez-Piernas et al. (2019) revealed the presence of 17 OMCs in leaves and 8 in fruits with a higher frequency of detection of carbamazepine, evidencing their higher capability of uptake and translocation within the plant. Sunyer-Caldú et al. (2023) found that pharmaceuticals were the most frequently detected ECs in soils and waters, whereas UV filters achieved the highest concentrations. Diclofenac and salicylic acid were the most accumulated in soils, and diclofenac, ofloxacin, and benzophenone-4 were the most prevalent in the WWTP effluent. Camacho-Arévalo et al. (2021) analyzing the fate of sulfonamide antibiotics in tomato crops in commercial greenhouses in Almería (Spain) found that sulfamethoxazole was the antibiotic with the highest concentration in tomato fruit and irrigated soils. Christou et al. (2017) in a long-term (three consecutive years) wastewater irrigation of a tomato crop found that the highest soil concentration was due to sulfamethoxazole whereas diclofenac displayed the highest fruit concentration. The concentration of the studied pharmaceuticals in both the soil and tomato fruits varied depending on the qualitative characteristics of the treated effluent applied and the duration of WW irrigation. EC concentrations in irrigation water, as well as their physiochemical properties (primarily charge and lipophilicity), are the primary determinants of their translocation and accumulation in the soil-plant continuum (Ben Mordechay et al., 2022).

The mass balance of the 10 ECs presented in this study was computed using the lysimetric technique utilized in this investigation for the soil, plant, and water compartments. Tables 3 and 4 indicate the total ECs intake in the systems (lysimeter) via irrigation water (90 and 270 mg lysimeter^-1 of each EC, respectively, plus the amount present in the freshwater); the same tables also show the number of ECs detected in plants, leached water, and soil in the TWWx1 and TWWx3 treatments. The not detected column is the residue of the mass balance between ECs intake and the measured sum of ECs accumulated in plants, leached water, and soil.

According to the mass balance, no ECs were found in the FW treatment; however, contaminants accumulation in the soil-plant-water system was measured for some ECs in the TWWx 1 ( Table 3 ) and TWWx 3 ( Table 4 ) treatments, with varying behavior among the ECs. Naproxen and diclofenac were not found in the plant tissues, soil, or drainage water of any of the irrigation treatments ( Tables 3 , 4 ). This means that nearly all of these ECs are degraded in different chemical by-products. Ketoprofen behaved similarly to naproxen and diclofenac, except for a 1% accumulation in the soil in the TWWx3 treatment ( Table 4 ).

Climbazole, clarithromycin, trimethoprim, metoprolol, and sulfamethoxazole accumulated in the soil as a percentage of the total amount of irrigation added to the system, with values ranging from 100%, 47%, 13%, 11%, and 4% in TWWx1 to 91%, 75%, 16%, 31%, and 6% in TWWx3 ( Table 3 , 4 ). Except for climbazole (1% in TWWx3) and sulfamethoxazole (3% and 8% in TWWx1 and TWWx3) in drainage water, no accumulation of these five ECs was detected in plant tissues or leached water. We assume that naproxen and diclofenac were degraded in by-products because the residual amount of these five ECs concerning total intake was not detected.

Fluconazole and carbamazepine were found in the soil, plant tissues, and drainage water. Carbamazepine accumulated in plant tissues, drainage water, and soil at a rate of 3%, 1%, and 49% of the total amount added to the system with irrigation in TWWx1 and 4,5%, 4%, and 39% in TWWx3. The balance that was not detected (47% and 53% in TWWx1 and TWWx3) is assumed to be degraded in by-products ( Tables 3 , 4 ). Fluconazole accumulated in plant tissues, drainage water, and soil at rates of 2%, 14%, and 58% in TWWx1 and 2.5%, 17%, and 70% in TWWx3. The balance’s undetected residual (26% in TWWx1 and 11% in TWWx3, respectively) is assumed to be degraded in by-products ( Tables 3 , 4 ).

Table 5 shows the average EC concentrations in tomato fruits. All fruits’ concentrations are given in fresh weight, with a ripe tomato containing 95% water and 5% dry matter. The results showed that the contaminants under study had varying concentrations and behaviors. None of the ten contaminants evaluated were discovered in significant concentrations in FW or TWWx1-irrigated tomatoes ( Table 4 , Figure 5 ). Some contaminants responded differently after TWWx3 treatment ( Table 4 , Figure 6 ). During the TWWx3 strategy, only fluconazole, carbamazepine, metoprolol, clarithromycin, climbazole, and sulfamethoxazole were identified in fruits. The concentrations of the individual compounds varied significantly: fluconazole was 110 ng g^-1, carbamazepine was 89.2 ng g^-1 and metoprolol was 1.22 ng g^-1. Clarithromycin, climbazole, and sulfamethoxazole were found at 0.03 ng g^-1 concentrations, which was statistically comparable to 0. Christou et al. (2017) discovered that diclofenac, sulfamethoxazole, and trimethoprim concentrations in soil were 0.35, 0.98, and 0.62 μg kg⁻¹, respectively. For fruit, diclofenac, sulfamethoxazole, and, trimethoprim concentrations were 11.63, 5.26, and 3.4 μg kg⁻¹, respectively. The average carbamazepine content in tomato leaves was 8.9 ng g⁻¹ while in fruit was 0.23 ng g⁻¹ (Martínez-Piernas et al., 2019). In tomato mature plants grown on fortified water-irrigated plots, the concentration of carbamazepine was found to be 0.19 ± 0.32 ng g^-1 (Wu et al., 2014). Ben Mordechay et al. (2022) discovered that the average EC content in soils was 129.4 88.5 g ha^-1, whereas the concentration of carbamazepine on tomato leaves was 546.4 557.5 ng g^-1.