1. Introduction
Phytoplankton has a major influence on marine ecosystems because of its role in oxygen production and overall significance in supporting and maintaining marine life. Due to the challenges in the direct measuring of phytoplankton, chlorophyll
a is mainly used as a proxy for phytoplankton biomass [
1,
2,
3,
4,
5]. Traditional techniques for chlorophyll analysis are based on bottle sampling, which is limited in terms of obtaining a high spatial and temporal resolution. This was corrected by modern in situ fluorescence techniques and remote sensing observations, which allow for measurements of chlorophyll in a high resolution. The research on chlorophyll temporal and spatial variability in the Mediterranean Sea is mostly restricted to the surface, as most of the collected data come from satellite observations and ex-situ laboratory measurements [
5]. The oligotrophic status of the Mediterranean basin [
6], with a decreasing trend from west to east [
7,
8], and seasonal and interannual variability [
4,
9,
10] were observed based on surface chlorophyll analysis.
The vertical distribution of chlorophyll is under the influence of many biotic and abiotic processes, whose interplay affects the depth of the upper mixed layer and controls the vertical chlorophyll distribution [
4,
11]. So far, information about vertical chlorophyll profiles has been derived from different kinds of measurements [
12,
13,
14,
15,
16,
17,
18,
19,
20,
21,
22,
23].
A similar chlorophyll seasonality was noticed in the Adriatic Sea by remote sensing observations [
24,
25,
26] and laboratory measurements [
27,
28]. The Adriatic Sea is a semi-enclosed basin of the Mediterranean Sea, where the biogeochemical parameters are influenced mainly by the circulation of the water masses. The chlorophyll trends in the Adriatic Sea have been investigated by remote sensing [
29,
30,
31] and laboratory-based measurements [
27,
32] to find deep chlorophyll maxima [
28,
33] and to investigate the anthropogenic impact on the phytoplankton community structure and abundancy [
34,
35,
36].
Coastal waters are highly sensitive ecosystems and may become even more vulnerable when exposed to the negative impact of human activities. Wastewater effluents contain high concentrations of pollutants, which can cause a large amount of deterioration to marine environments [
37]. Bacteria are the most abundant and diverse group of organisms living in wastewater [
38,
39], and bathing water status is mainly determined using fecal indicator bacteria (FIB) as indicators of water quality and related human health issues [
40,
41]. Besides fecal pollution, eutrophication is another important aspect of coastal water quality. Since primary producers are the first components of an ecosystem to react to an increase in nutrient concentration, phytoplankton biomass is considered a good indicator of the trophic state of coastal water [
42]. Changes in phytoplankton biomass abundance occur following nutrient enrichment, which can happen naturally (through riverine influence or the upwelling of nutrients from the bottom) or under anthropogenic influence (urban and industrial wastewater discharges and surface runoff). Anthropogenically introduced nutrients are the major problem of coastal ecosystems [
43,
44], and their mixing with seawater contributes to the eutrophication of coastal zone waters [
45], thus increasing the biomass of macroalgae and seston algae [
46].
This paper aims to characterize chlorophyll vertical profiles and describe their spatial and temporal variability based on the stratification status of a water column and the presence of a marine wastewater effluent plume. Due to the nature of fluorescence measurements, which change with the taxonomic species or environmental conditions [
47] and can be affected by dissolved materials [
48] or by different light irradiation statuses [
49,
50,
51], the present study is focused on the relative distribution of vertical chlorophyll fluorescence in a water column by comparing different segments of the water column for each considered profile. To our knowledge, this is the first classification of chlorophyll vertical profiles in the presence of a wastewater effluent plume. The spatiotemporal variability of the different vertical profiles concerning stratification was also investigated.
2. Materials and Methods
2.1. Study Site
The study area is the eastern central Adriatic Sea, near the city of Split in Croatia (
Figure 1). Two main marine wastewater outfalls are located within the study area, about 5 km away from each other, as a part of the Split–Solin public sewage system.
The submarine outfall at Location 1 discharges treated wastewater from the Katalinića brig wastewater treatment plant (WWTP) via a Ø 800 mm pipeline, ending with two diffuser sections 1.3 km away from the shore at a 43 m depth. The Katalinića brig WWTP is designed for the mechanical treatment of both municipal and rainfall-runoff wastewater, with a capacity of 122.000 population equivalents (PE) and an average flow rate equal to 35.000 m3/day approximately.
The marine outfall at Location 2 (Stobreč) discharges the wastewater coming from the Stupe WWTP, where the wastewater is mechanically treated, along with grit and grease removal. The Stupe WWTP facility is designed for municipal wastewater only, with a capacity of 138.000 PE and an average flow rate equal to 30.000 m3/day. The length of the submarine outfall Ø 900 mm pipeline is 2.76 km, with a 200 m long diffuser section ending at a 37 m depth.
Additionally, the monitoring of different physical and chemical parameters is performed several times a year, at stations FP-O14 and FP-O14b, as shown in
Figure 1, as a part of the national monitoring program.
To understand the background of the wastewater being discharged into the seawater, we provide data (
Table 1) about the seasonal flowrate and available water effluent parameters for two WWTPs.
2.2. Sampling Methodology
This study was performed during the implementation of the Interreg Italy–Croatia project AdSWiM (Managed use of treated urban wastewater for the quality of the Adriatic Sea). The measurements were performed with the CTD fast profiling probe (The Sea & Sun Technology) and C3 fluorometer during sampling sessions conducted with a boat (Nautica 600) every month from February to September 2020 (with the exception of March and August due to the COVID restrictions).
CTD fast profiling probe, with an additional sensor for dissolved oxygen (model: 48M), was used for temperature, salinity, density anomaly, and dissolved oxygen measurements. The temperature is measured by a classical resistance thermometer using a platinum resistor, while the salinity is calculated from the seawater conductivity using a 7-poll cell sensor. The density anomalies were calculated automatically by the CTD/DO probe internal software using the international thermodynamic equation of seawater-2010 [
52]. The dissolved oxygen sensor works on the principle of a Clark electrode. The depth is calculated from the pressure, measured by a strain gauge, and is converted to depth as 1dbar = 1 m. The turbidity, chlorophyll, and colored dissolved organic matter (CDOM) were measured using a C3 Submersible fluorometer (Turner Design). The chlorophyll measurements were performed with a blue mercury lamp, with a peak emission at 460 nm, and the fluorescence was collected at 680 nm. The CDOM was measured using a UV LED (CWL: 365 nm), with a peak emission at 350 nm and fluorescence collection at 450 nm. The turbidity was measured with the IR lamp, with a peak emission at 850 nm, and the collection of scattered light at 90°. Since the relationship between phytoplankton biomass and chlorophyll fluorescence is highly variable and depends on the physiological state and community composition [
53,
54,
55], in this study, we focused on the general distribution of chlorophyll profiles in the water column concerning the stratification and effluent discharge.
The fecal pollution was estimated by the level of FIB. The FIB concentration was assessed by the membrane filtration method and the enumeration of fecal coliforms (
E. coli) [
56,
57] and intestinal enterococci (ENT) [
58], as indicated in the Croatian legislation (Regulation on the Quality of Marine Bathing Waters; OG 73/08). The number of viable heterotrophic bacteria was determined using the spread plate method. An automated Protos3 colony counting system (Synbiosis, UK) was used to determine the bacteriological concentration. Sampling was conducted from the boat using a Niskin sampler. The samples were collected in sterile bottles, protected from light in cool boxes, and transported to the microbiology laboratory of the Faculty of Science in Split to be processed within 4 h.
The chlorophyll concentration data at locations FP-O14 and FP-O14b were collected during the regular monitoring performed by the water management agency (Croatia Waters). The chlorophyll a concentrations were determined by the laboratory fluorometric method, with filtration through Whatman GF/C glass filters, at 0, 5, 10, 20, and 50 m depths at the FP-O14 location and at 0, 5, 10, and 43 m depths at FP-O14b.
2.3. Data Analysis
The data collection was performed by the vertical profiling of the water column down to 40 m and 35 m depths at location 1 and location 2, respectively. A moving average filter was applied to raw data sampled at a maximum spatial resolution of 3 cm to achieve spatial uniformity. The data were smoothed by a moving average filter, with a 20 cm window.
The vertical distributions of the different layers were examined by calculating the vertical gradients of the density anomalies and derived buoyancy frequency. The buoyancy frequency or Brunt-Väisälä frequency (N) represents the strength of the density stratification and, consequently, the stability of the water column. It is defined as:
where g is the gravitational acceleration,
0 is the averaged density of the whole water profile, and d
/dz is the water density gradient at each sampling step. To identify events when stratification occurred, data points are selected where N
2 exceeds 0.001 s
−2 as possible pycnocline candidates [
59]. The data points are then aggregated if the distance between them was less than 0.8 m, forming a possible pycnocline layer. Layers thinner than 0.8 m are neglected to exclude the impact of anomalous extremes. Aggregated layers are classified as a pycnocline and their profiles as stratified, which potentially aids in the creation of chlorophyll layers. To support the analysis of forming chlorophyll layers under the influence of marine wastewater discharge, the vertical profiles of dissolved oxygen, turbidity, CDOM, temperature, and salinity are also examined.
4. Discussion
The majority of the vertical chlorophyll profile studies conducted in coastal region and estuaries are based on laboratory measurements or satellite techniques, lacking direct in situ measurements, especially the influence of marine outfalls. Therefore, this study aimed to identify the chlorophyll vertical profiles under the influence of the water column stratification status and the presence of an effluent plume. Based on the measurement results, we classify the chlorophyll profiles into three types (
Table 2). The profiles observed in February and April 2020, with the observed rise of the chlorophyll concentrations from the bottom to the top of the water column, are classified as the Upper Chlorophyll Profile (UCP). The other two profile types are divided between the sampling sessions when stratification of the water column was observed, and they differ by the presence of an effluent plume. In May at both locations and in September at location 2, the vertical chlorophyll profile was characterized by a higher concentration near the bottom, without a decrease in dissolved oxygen. These profiles are classified as the Bottom Chlorophyll Profile (BCP). The third profile type is characterized by peaks of chlorophyll coinciding with CDOM and turbidity peaks and a matching decrease in salinity and dissolved oxygen, which is a consequence of an effluent plume intrusion. These profiles are classified as the Effluent Chlorophyll Profile (ECP). A summary of this classification is given in
Table 2.
In unstratified conditions, observed in February and April 2020, at both locations, the upwelling of the bottom, nutrient-rich waters induced a rise in the chlorophyll concentrations from the bottom to the top of the water column, which we classified as UCP. This rise is supported by the CDOM and turbidity vertical profiles, which show only small variations along the vertical column. The bottom water upwelling also resulted in a higher FIB level at the surface, indicating a rise in fecal pollution due to marine outfalls. In unstratified conditions, the dissolved oxygen vertical profiles were homogeneous. This is due to vertical mixing, which allows air–sea gas exchange with the bottom waters and the influence of external forces, such as wind [
61]. An exception occurred in April 2020 at location 1 due to an effluent plume intrusion, which decreased the dissolved oxygen values. These conditions induce the characteristics of the chlorophyll profile of both the ECP and UCP (
Figure 2). The oxygenation of the water column down to a 30 m depth was due to the primary production of phytoplankton, but it was also due to the unstratified water column, which caused vertical mixing and aeration of the entire water column. At the bottom of the water column, the dissolved oxygen concentration and chlorophyll are inversely correlated, suggesting that the consumption of oxygen dominates over its production. Similar observations have been made in many estuaries with nutrient inputs from rivers [
62,
63,
64,
65,
66,
67]. Marine outfalls discharge nutrients into the sea, resulting in similar biological processes occurring as those when rivers discharge nutrients into the estuary.
From May to September, stratification occurred at both locations, and the observed profiles are classified as BCP (May 2020, September 2020) and ECP (June 2020, July 2020, September 2020) (
Table 2). BCPs are characterized by a higher chlorophyll concentration near the bottom of the water column. This increase is likely due to the phytoplankton activity and resuspension of the bottom sediment, where dead plankton and other sources of organic matter fertilizer, unconsumed food, and fecal matter settle. In May 2020, despite the water column stratification, the vertical distribution of dissolved oxygen was homogeneous, similar to a phenomenon observed in some estuaries, where, under stratified conditions and higher temperatures, hypoxia occurred [
68,
69,
70]. In our case, hypoxia was never observed, even in summer, but a higher dissolved oxygen variability was present.
To confirm the occurrence of UCP and BCP types in the absence of an effluent plume, we used data from measurements conducted by the Croatian water management agency in the period from 2012–2019, sampled at two locations (FP-O14 and FP-O14b). Both stations are used to represent coastal seawater unaffected by marine outfalls. At both of these locations, during months when no stratification is present, the vertical chlorophyll profile shows upwelling, indicating the existence of UCP (
Figure 6 and
Figure 7). Under stratified conditions, narrower ranges of the chlorophyll concentrations are observed in the upper part, with the highest values measured at the bottom, which are classified as BCP (
Figure 6 and
Figure 7). These data show the existence of UCP and BCP profiles in coastal, unpolluted waters, which follow a seasonal cycle of sea warming and the formation of stratified conditions, and differentiate the naturally occurring chlorophyll profiles from the ones induced by anthropogenic influence.
The vertical profiles of CDOM, turbidity, dissolved oxygen, and salinity further highlight the difference between ECP and BCP. The nutrient enrichment by marine outfalls induces phytoplankton biomass growth and increases the chlorophyll concentration. The highest chlorophyll concentration observed was at location 1 in July 2020. Such a surge in the chlorophyll concentration could be due to the accumulation of living algae, supported by raw sewage, containing higher amounts of nutrients, especially nitrogen and phosphorus. Besides nutrients, the marine outfalls also discharge organic matter into the coastal waters, resulting in increased organic matter concentrations and turbidity. The increased turbidity levels are the result of the input of organic matter from the wastewater discharge, creating a good condition for the presence of fecal bacteria. The effluent intrusion creates conditions where the oxygen consumption, during the bacterial oxidation of organic matter and decomposition of dead algae and plants, exceeds the production of dissolved oxygen by photosynthesis. The observed decrease in the dissolved oxygen profiles furthermore separates the BCP and ECP types. The gradient of dissolved oxygen with depth (higher concentration above the pycnocline) indicates that stratification may play a role in lowering the oxygen concentration by reducing the gas exchange between the air and sea surface layer [
61]. While stratification can influence the dissolved oxygen profile, the homogeneously dissolved oxygen profiles observed in May suggest that the highest variability of dissolved oxygen concentration is due to the influence of a wastewater effluent plume and not stratification. The variations in salinity also supported our hypothesis about the nature of ECP. Similar changes in salinity were previously observed for other submarine outfall sites and provide further evidence of a freshwater input through the outfall system [
71,
72,
73,
74].
Due to the fact that both WWTPs are discharging a municipal wastewater effluent, the seasonal variations are due to the variations in the residential population of the Split municipal area, which increases during the summer months. Consequently, such changes can influence the wastewater volume. Thus, the summer discharges are approximately 12% higher than those during the winter period, as presented in
Table 1. This is also shown in
Figure 2 and
Figure 3 in terms of an increased concentration and/or gradients of measured parameters. While the discharged wastewater and the seawater start mixing quickly and reaching equilibrium with the surrounding seawater, the salinity profiles may serve as additional information, aiding in distinguishing the chlorophyll profiles. During the observed stratification, the effluent plume was held below the pycnocline, which is generally ecologically favorable, as it maintains a higher quality of the surface waters.
In addition, the circulation in the wider area surrounding the two studied outfalls has a typical barocline current pattern, with relatively larger surface layer velocities due to the regular wind intensities. Furthermore, during the unstratified conditions, the bottom upwelling is intensified, creating the potential for effluent plume surfacing. In such cases, it is expected that surface currents will further contribute to the mixing and spreading of chlorophyl.
5. Conclusions
To study the wastewater impact on the chlorophyll vertical distribution, measurements were performed in the eastern central Adriatic Sea at two locations of marine outfalls, near the city of Split in Croatia. We describe the three vertical chlorophyll profile types, based on the stratification conditions and the effluent plume presence. The water column mixing induced the upwelling of the bottom water, generating UCP and enabling the rise of the effluent towards the surface. In the stratified conditions, categorized as the ECP type, the pycnocline prevents the plume from rising, keeping it in the lower part of the water column, while the BCP type of the profile represents a case when the water column is stratified, but no effluent plume is present. Our categorization of the different vertical chlorophyll profiles is supported by data collected by the Croatian water agency at the locations unaffected by the marine outfalls. All three observed chlorophyll profile types are the consequence of the interplay between the physical, biological, and chemical processes. To distinguish these processes and confidently identify the presence of the effluent plume we use the FIB measurements, along with dissolved oxygen gradients, to examine the conditions when the oxygen consumption exceeds the production, indicating organic matter oxidation. This, together with the salinity, CDOM, and turbidity measurements, allows us to separate the wastewater influence from the natural processes. We confirm the importance of the stratification in submerging the effluent below the pycnocline, which is highlighted in February 2020, when column mixing allowed the FIB to rise to the surface of the water column. To our knowledge, this is a first study involving in situ vertical measurements coupled with the submarine outfall, analyzing its influence on the vertical chlorophyll distribution.