Impact of Depuration Plants on Nutrient Levels in the North Adriatic Sea

: Macronutrients (nitrogen—N; phosphorus—P; silicon—Si) play a crucial role in ocean surface waters stimulating the planktonic primary production; in fact, their concentrations are fundamental for the evaluation of the trophic status of the water body and eutrophication phenomena. Loads of nutrients into the sea are mainly represented by river runoff and depuration plant outﬂows. For this purpose, in the framework of the AdSWiM project, “Managed use of treated urban wastewater for the quality of the Adriatic Sea” levels of N-NO 3 , N-NO 2 , N-NH 4 , Si-Si(OH) 4 , P-PO 4 (dissolved inorganic phosphorus—DIP) and total dissolved phosphorus (TDP) were determined colorimetrically at two sites in the Gulf of Trieste: Lignano Sabbiadoro and San Giorgio di Nogaro. For each site, during the bathing seasons of 2019 and 2020, a sample from the depuration plant (DP) outﬂow and another one in the bottom seawater near the discharging pipelines were collected. Results showed a strong dilution effect on nutrient levels passing from DPs to the sea, from one to three orders of magnitude and a low and not harmful concentration in seawater. The outﬂow composition of the two DPs showed that the main fraction of dissolved inorganic nitrogen (DIN) was represented by N-NO 3 for Lignano, while in San Giorgio the major contribution came from N-NH 4 . Concerning phosphorus, Lignano showed a higher content (about 3 times) of P levels than San Giorgio, but a similar percentage composition, DIP:DOP (77:23), compared to the seawater site one DIP:DOP (2:98). Despite the difference between the DPs, no substantial differences were found in the sea sites, demonstrating the negligible effect of the DP outﬂows in the nutrient levels in the study area.


Introduction
Eutrophication is an environmental condition of degradation generated by excess nutrient levels (mainly nitrogen and phosphorous) in seawater or freshwater that can produce a series of problems such as algal blooms, anoxia-as a consequence of excessive oxygen consumption-and increased biological degradation processes, resulting in modifications of benthic communities and fish mass mortality events [1][2][3].
Eutrophication is listed as a Descriptor in the Marine Strategy Framework Directive [4] for the definition of the marine waters' Good Environmental Status (GES) as "Humaninduced eutrophication is minimized, especially adverse effects thereof, such as losses in biodiversity, ecosystem degradation, harmful algae blooms and oxygen deficiency in

Sampling Stations
Two types of samples were collected: (1) "DP-Depuration Plants", which consisted of treated wastewater collected at the end of the water depuration processes and (2) "Sea", sampled at the sea bottom next to the respective wastewater discharge point. Two sites with associated DP-Sea stations were selected: Lignano (Lignano DP: 13.107832 • E, 45.  Samples were collected monthly during two bathing seasons (2019 and 2020) from April to October (no samples were collected during September) [29]. DP samples of treated sewage were taken before unloading, just upstream from the injection into the discharging pipeline, filtered immediately through GF/F filters and collected directly in 50 mL decontaminated PET centrifuge tubes. Seawater samples were collected in the lowest layer of the water column (depth 13.7 m) in the proximity of each DP outfall (1 m above the main diffusion point of the pipelines) by means of 5 L Ruttner bottles, and then filtered and collected as described for DP samples [28]. All tubes were stored at −20 • C until being processed in the laboratory.

Laboratory and Apparatus
All sample treatments and analyses were carried out in a clean room laboratory ISO 14644-1 Class 6, with areas at ISO Class 5 under laminar flow. The acid-cleaning procedures, used for all laboratory materials, were performed as described by [30,31] with HCl purchased from Carlo Erba, Milan, Italy. A two-stage system Midi (Elix and Milli-Q) from Millipore (Bedford, MA, USA) was used to produce ultrapure water. All reagents used for the preparation of standards and working reagents were purchased by Merck, Italy. Variable volume micropipettes and neutral tips were from Brand (Wertheim, Germany, Transferpette).

Nutrient Analyses
An EasyChem Plus (SYSTEA s.p.a., Frosinone, Italy) automated discrete analyzer set up with standard colorimetric methods was used to determine nutrient levels using colorimetry. Three replicates for each sample were analyzed, SD and RSD were calculated to enhance the accuracy of the measurements. The calibration curve method was used for quantification and for each nutrient. Instrument LOD and LOQ were evaluated according to ICH Q2B (ICH, 2005). A washing cycle with a 1:1000 HCl super pure solution was applied between each analysis in order to avoid cross-contamination phenomena.
For N-NO 3 , Ref. National Environmental Methods Index 9171 Nitrate via V(III) reduction was used. Nitrates are reduced to nitrites by an acid solution of vanadium chloride and then determined as nitrite. This vanadium-based method directly measures the nitrate concentration, since the nitrite fraction is subtracted automatically as part of the sample blank (LOD = 0.0663 µmol L −1 ; 0.9286 µgL −1 ; LOQ = 0.201 µmol L −1 ; 2.814 µgL −1 ).
N-NH 4 was determined with the EPA #350.1 method. Ammonia reacts in basic conditions with sodium salicylate and hypochlorite in the presence of nitroprusside salts to form an emerald-green chromatic substance. The pH in the colorimeter cell must approach 12.6. Colorimetric measurements were performed at 670 nm (LOD = 0.08 µmol L −1 ; 1.45 µg L −1 ; LOQ = 0.244 µmol L −1 ; 4.41 µg L −1 ).
Organic P was evaluated by the difference between total dissolved P (TDP) and inorganic P. Total P was obtained by adding the oxidant mix of persulfate, boric acid, and sodium hydroxide and the samples were placed in an autoclave at 120 • C for 45 min as described by [32]. TDP was then immediately determined with the EPA #365.1 method as PO 4 3− . For Si-Si(OH) 4 , the APHA Standard Methods for the Examination of Water and Wastewater 4500-Si(OH) 4 method was used. Silicates were mixed with an acid solution of ammonium molybdate to produce Si[Mo 12 O 40 ] 4− . The measurement was performed in the presence of oxalic acid to mask the interference of phosphates and the Si-anion was then reduced by ascorbic acid to form the blue-colored β-Keggin ion derivative. Colorimetric measurements were performed at 880 nm (LOD = 0.0789 µmol L −1 ; 2.2 µg L −1 ; LOQ = 0.2392 µmol L −1 ; 6.71 µg L −1 ).
All analyses were performed between 12 and 36 h after the sample thawing, SD and RSD were calculated to enhance the accuracy of the measurements. An accuracy test was performed using certified reference material "QC3179-Simple nutrients in seawater", purchased from Merck, processed during the analysis as the quality control.
Since we only had punctual values, nutrient levels found in the depuration plants only count as indicators for the study and they have no legal value due to the different sampling method from the one described in the Dlg.s 152/2006 [33].

Statistical Analysis
Data are expressed as arithmetic mean ± standard deviation (SD) of the performed replications (n = 3). Statistical analyses were performed using the analysis of variance (one-way ANOVA), followed by the Multiple range test, after testing the homogeneity of the variance with Levene's test. In case the data did not show a homogeneous variance, the Kolmogorov-Smirnov non-parametric test (for comparison between two groups) or the Kruskal-Wallis test (for comparison between three or more groups) was applied. Significant differences were evaluated at the 95% confidence level.

Seawater
The levels of N-NO 3 ranged from a minimum of 0.28 ± 0.01 (Lignano Sea, May) to a maximum of 7.59 ± 0.03 µmol L −1 (Lignano Sea, April), with a mean of 2 ± 2 µmol L −1 in 2019 and from a minimum of 0.44 ± 0.04 µmol L −1 (San Giorgio Sea, April) to a maximum of 2.40 ± 0.09 µmol L −1 (San Giorgio Sea, August) with an average of 1.6 ± 0.6 µmol L −1 (Figure 2a) in 2020. For N-NO 3 , no particular trend was evident, except for a general small increment until August and a diminution in October, with no differences between the sites, except for a peak in Lignano, May 2019. With respect to the N-NO 2 levels, a range from a minimum of 0.40 ± 0.02 µmol L −1 (San Giorgio Sea, July) to a maximum of 0.82 ± 0.08 µmol L −1 (San Giorgio Sea, August) with a mean of 0.5 ± 0.1 µmol L −1 during 2019 and from a minimum of 0.22 ± 0.01 µmol L −1 (San Giorgio Sea, May) to a maximum of 0.77 ± 0.03 µmol L −1 (San Giorgio Sea, June) with a mean of 0.4 ± 0.2 µmol L −1 in 2020 ( Figure 2c). Differently from 2019, for which a bell-shape trend was evident for both sites, in 2020 there was no particular trend. The range for N-NH 4 was from a minimum of 0.40 ± 0.03 (San Giorgio Sea, May) to a maximum of 2.5 ± 0.2 µmol L −1 (Lignano Sea, June) with a mean of 0.9 ± 0.6 µmol L −1 in 2019. In 2020, N-NH 4 ranged from a minimum of 0.50 ± 0.01 µmol L −1 (San Giorgio Sea, May) to a maximum of 18.3 ± 0.7 µmol L −1 (Lignano Sea, April) consisting of a peak of one order of magnitude higher than all the other samples, treated as an outlier in the subsequent results with a mean value of 2 ± 1 µmol L −1 (Lignano Sea, April, excluded from mean and dev.st) or 3 ± 5 µmol L −1 (Lignano Sea, April, included) ( Figure 2e). A bell-shape trend with a peak in June is evident for every site.
In 2019, most of the seawater DIP values were below LOD (0.01 µmol L −1 ), with a maximum of 0.03 ± 0.01 (San Giorgio Sea, July) and a mean value of 0.015 ± 0.01 µmol L −1 . Even in 2020, DIP levels were below the LOD in most cases, with a maximum of 0.05 ± 0.01 µmol L −1 (San Giorgio Sea, May) and a mean value of 0.02 ± 0.01 µmol L −1 (Figure 2i). DIP levels generally increase during the middle of the season. TDP ranged from a minimum of 0.36 ± 0.02 µmol L −1 (San Giorgio Sea, October) to a maximum of 1.6 ± 0.1 µmol L −1 (San Giorgio Sea, July) with a mean value of 0.8 ± 0.4 µmol L −1 in 2019 and from a minimum of 0.43 ± 0.04 µmol L −1 (Lignano Sea, April) to a maximum of 2.3± 0.3 µmol L −1 (San Giorgio Sea, April) with a mean value of 1.4 ± 0.6 µmol L −1 in 2020 ( Figure 2k). No significative differences were found between the sites. TDP concentrations tended to increase during the middle of the season followed by a decrease at the end.

Depuration Plants
In DPs, N-NO 3 levels ranged from a minimum of 8.3 ± 0.7 µmol L −1 (San Giorgio DP, April) to a maximum of 445 ± 1µmol L −1 (Lignano DP, August), with a mean of 192 ± 178 µmol L −1 in 2019 and from a minimum of 0.15 ± 0.01 µmol L −1 (San Giorgio DP, July) to a maximum of 504 ± 43 µmol L −1 (Lignano DP, June) with an overall mean of 143 ± 176 µmol L −1 (Figure 2b) in 2020. Lignano DP data showed a bell-shape trend both in 2019 and 2020, while for San Giorgio, which always displayed lower values, there was no particular trend. N-NO 2 levels ranged from a minimum of 0.98 ± 0.1 µmol L −1 (San Giorgio DP, October) to a maximum of 59 ± 2 µmol L −1 (Lignano DP, May) with a mean of 11.1 ± 19.1 µmol L −1 during 2019. In 2020 DPs, N-NO 2 levels ranged from a minimum of 0.400 ± 0.004 µmol L −1 (San Giorgio, October) to a maximum of 86.0 ± 2.5 µmol L −1 (Lignano DP, June) with a mean of 11 ± 24 µmol L −1 (Figure 2d). Nitrite levels in 2020 were significantly higher than 2019, although no trend was evident for both years. N-NH 4 ranged from a minimum of 0.4 ± 0.2 µmol L −1 (Lignano DP, August) to a maximum of 83 ± 3 µmol L −1 (San Giorgio, June), with a bell-shape trend and a mean of 17 ± 23 µmol L −1 during 2019 and from a minimum of 1.4 ± 0.04 µmol L −1 (Lignano DP, October) to a maximum of 295 ± 1 µmol L −1 (San Giorgio DP, April), with a mean value of 115 ± 97 µmol L −1 in 2020. Despite the peak in April, N-NH 4 in Lignano DP followed a bell-shape trend as seen in 2019; San Giorgio showed more variable data, with a minimum in July. San Giorgio DP showed for both 2019 and 2020 a higher value than Lignano DP (Figure 2f). The concentration of Si-Si(OH) 4 ranged from a minimum of 87 ± 1 µmol L −1 (San Giorgio DP, October) from a maximum of 159 ± 2 µmol L −1 (San Giorgio DP, June) with a mean value of 125 ± 44 µmol L −1 in 2019 and from a minimum of 26 ± 1 µmol L −1 (San Giorgio DP, August) to a maximum of 254 ± 5 µmol L −1 (San Giorgio DP, July) with a mean of 101 ± 72 µmol L −1 in 2020 (Figure 2h). While in 2019 no trend was evident, during 2020 silicate concentrations followed a bell-shape trend.
In DP samples, DIP ranged from a minimum of 4.2 ± 0.45 µmol L −1 (San Giorgio DP, May) to a maximum of 57 ± 3 µmol L −1 (Lignano DP, October) with a mean value of 21 ± 19 µmol L −1 in 2019 and from a minimum of 1.4 ± 0.1 µmol L −1 (San Giorgio DP, October) to a maximum of 73.1 ± 0.3 µmol L −1 (San Giorgio DP, July) with a mean value of 19 ± 23 µmol L −1 in 2020 (Figure 2j). The average contribution for DIP to TDP in DP samples was 72% during 2019, while in 2020 it was 81.5% of the total (Figure 4). No particular trend was evident. For TDP, the concentration varied from 6.4 ± 0.4 µmol L −1 (San Giorgio DP, April) to a maximum of 77 ± 3 µmol L −1 (Lignano DP, October) with a mean value of 9 ± 8 µmol L −1 in 2019 and from a minimum of 4.0 ± 0.2 µmol L −1 (San Giorgio DP, October) to a maximum of 79 ± 1 µmol L −1 (San Giorgio DP, July) with a mean value of 23 ± 23 µmol L −1 in 2020 (Figure 2l). No particular trend was evident for Depuration Plants TDP concentrations.
The contribution to TDP incoming from DIP or DOP is strongly comparable between DP and seawater. Between 2019 and 2020, DIP % increased in both DP sites: Lignano DP from 75% to 81.4% and San Giorgio DP from 68.4% to 81.5%. The same increment was not found in seawater, which remained almost constant between the sites and the years (average 2% DIP) (Figure 4).
In 2019, the DIN:DIP ratio ranged from a minimum of 62 (San Giorgio Sea, July) to a maximum of 728 (Lignano Sea, April), with a mean of 235 ± 191. In 2020, DIN:DIP ranged from a minimum of 104 (San Giorgio Sea, April) to a maximum of 1771 (Lignano Sea, April), with a mean of 353 ± 458. During 2019, there were no significant differences between the sites (p > 0.05), average values Lignano 256 ± 238 and San Giorgio 214 ± 151. During 2020, we observed a statistically significant difference (p < 0.05) between the sites for average values of Lignano (517 ± 626) and San Giorgio (188 ± 69) sea, principally ascribed to the high value of the Lignano April sample (mean without Lignano sea, April: 267 ± 133).

Discussion
In the present study, sampling stations were characterized by low depth (almost 15 m) and shared similar physical and geomorphological features. Lignano is a tourist spot and its population increases considerably in the summer season. Lignano DP is designed for the treatment of 70,000 inhabitants equivalent (i.h.) and treats mostly domestic sewage from the Lignano municipality, while San Giorgio DP is designed for the treatment of 800,000 i.h. but during the sampling activity actually treated 120,000 i.h., and differently from Lignano, the wastewater inputs are mostly industrial, coming from six different municipalities. In both DPs, in order to avoid a massive load of sewage discharge in a single spot, the outflow of the treated water takes place through a system of diffusers in a long section of the pipeline, at a depth of about 14-15 m.
The concentrations of nutrients determined in DPs waters and in the corresponding outflow sites at sea were highly different: a strong dilution factor between DP and Sea concentrations was determined for each parameter, from one to three orders of magnitude (Table 1). The dilution factor applied from the water body lowers the nutrient concentrations to ranges that do not result harmful to the environment and are comparable with the literature reported in Table 2. DIP levels in the seawater resulted very low, and in 66.6% of the cases below the detection limit (LOD = 0.011 µmol L −1 ). In fact, DIP is recognized as the limiting element for phytoplankton growth and, thus, for the trophic status in the study area as evidenced by experimental studies [34,35]. The similarity between the data determined at sea in this study and those reported in the literature for other offshore sites (Table 2) suggests that the nutrient levels might be majorly affected by the riverine inputs rather than by the sewage discharge in this area.
The DIP:DOP ratio within the treated wastewater is highly dependent on the type of effluent treated and the treatments applied. Only few studies characterized P fractions in different types of untreated wastewater. The values showed phosphate concentrations of 0.12-350 mg L −1 , accounting for 34-100% of TDP [36][37][38][39]. Other studies showed that domestic wastewater has a lower concentration of P compared to industrial, with concentrations typically varying from 5 to 30 mg L −1 depending on urban or rural wastewater [40][41][42][43][44]. About 78% of P is removed during the primary treatment; however, the treated effluent generally has a composition comprising a phosphate content of about 80-100% of the total P [45].
As for the values found in the sampling sites at sea, our results are confirmed by the study conducted by [46], where in the waters of the Gulf of Trieste the DOP always represents the dominant fraction (about 71% of TDP). TDP levels were generally higher than the reported literature, probably ascribed to the discharging pipeline's proximity. DOP represented 98% of the TDP at the Sea stations, with no differences between sites or years.
Over the two years, ammoniacal nitrogen (Figure 2e,f) increased its concentrations by about 3 times in Sea and 7 times in DP sites. This increment could be related to the difference in the sewage's composition while entering the DPs. Concerning DIN, domestic sewages, differently from the industrial ones, are mainly formed by N-NH 4 . The restriction induced by the COVID-19 pandemic event forced a suspension of the industrial activities from February to June-July 2020, which could result in a change in sewage composition, with less industrial and more domestic inputs for Depuration Plants. This increment in ammoniacal nitrogen also modified the DIN distribution, both in DPs and in Sea sites ( Figure S1a-c). Almost all the values found in the DPs resulted well below the Legal Limit [33], with two exceptions for nitrous nitrogen (Figure 2d). DP outflow nutrient concentrations are different both between the years and the sites. Concerning inorganic nitrogen, Lignano DP is mainly constituted by nitric nitrogen, while San Giorgio by ammoniacal nitrogen. Lignano DP generally has higher levels of nutrients, especially for nitrates, nitrites, and phosphorus. The main difference between the DP sites could concern the composition of the receiving sewage.
To ensure a more accurate comparison relative to our seawater sampling points at the depths we investigated (about 13.7 m), values of DIN, P-PO 4 , and Si-Si(OH) 4 were extracted from the EMODnet chemistry portal [47]. The results of the investigation are reported in Table 2 For the seawater nutrient levels, the comparison between our data and the literature showed comparable values for almost all the parameters (Table 2).
Concerning phosphorus, TDP and DOP are higher if compared to the studies carried out in the Gulf of Trieste and along the coast of Ancona [25,46,54], while the inorganic fraction is in line with the values recorded in the Adriatic basin.
Regarding nitrogen, DIN values measured in our study are similar to those recorded in [46], while they are higher than the bottom concentrations of the same site [47] and near the coast of Ancona [25]. The N-NO 2 values recorded in 2019 were generally lower than the other studies considered, while the 2020 values were comparable with similar studies carried out in the same sites during the first half of the 90s [48,49]. N-NO 3 and N-NH 4 fractions are consistent with the literature.
Finally, silicates are the only compounds with slightly higher concentrations than most of the published works, except for the sampling campaign carried out in 1992-1993 [51].

Conclusions
The present study is focused on the influence of treated wastewater plants on the open sea nutrient contents. The results of this study showed that (1) the depuration plants taken into consideration have a negligible impact on the marine ecosystem as the nutrients present in the treated wastewater undergo a strong dilution once they reach the marine basin. Phosphorus levels, both inorganic and total, are different in the DPs and are generally about 2.6 times higher in Lignano than in San Giorgio. Despite these differences in P levels, the distribution of the inorganic and organic fractions compared to total dissolved P is similar in the two DPs: the main fraction is always represented by DIP (average 77.2%), whereas DOP is about 22.8%. In the seawater, almost all of P is provided by DOP (average 98%). Seawater DIP levels were below LOD in most of the cases. There is a strong difference between Lignano and San Giorgio DP's outflow composition, probably due to the different typology of sewage inputs: concerning DIN, the minor fraction is represented in every case from N-NO 2 (from 1 to 5%) and Lignano DP is mainly constituted by N-NO 3 (94%-2019; 81%-2020), while San Giorgio DP from N-NH 4 (43%-2019; 88%-2020). On the other side, no significant differences were found in the sea site, underlining that the dilution factor is the driving force. Moreover, (2) no differences were identified between 2019 and 2020 except for ammoniacal and nitrous nitrogen. This study provided important information on the distribution of nutrients in seawater related to the activity of depuration plants. However, further studies are necessary to better understand the impact of discharges on marine basins in order to improve the knowledge on possible chemical, biological, and ecological implications. Data Availability Statement: All data taken into account for the study are published and available at https://nodc.ogs.it/catalogs/doidetails;jsessionid=C2C6BB7EFE1B44416357C25C8CBCE378?0 &doi=10.13120/j23k-n088 (accessed on 1 June 2022), https://doi.org/10.13120/J23K-N088. The dataset contains also Croatian parameters concerning depuration plants and seawater for the sites of Split and Zadar.