**5.1 Experiment I: June 18, 2009.**

The first experiment was carried out on June 18, 2009. The wave conditions were measured by the wave buoy located 5 nautical miles off the shore of Cesenatico (details on the wave position and data are available at http://www.arpa.emr.it/sim/?mare/boa). The significant

The results also reveal different effects on plume areal dispersion and on thermoaline profiles between zones confined by continuous breakwaters (north shore) and by discontinuous breakwater (south shore). Comparing salinity vertical distribution in internal and external points of the north continue breakwaters, under a surface layer (50-60 cm) almost corresponding to breakwater submergence (Lamberti et al., 2005), differences in salinity and oxygen profiles become significant. Freshwater dispersion appears obstructed in the internal north confined area because continuous breakwaters produce a "wall effect" for incoming plume with mass exchange reduced for deep layers. Here, in the absence of north directed sea currents, flows are allowed only from north-south boundary mouths with

**5. Validation of model results with in situ measurement campaigns** 

Simultaneously, tide, waves, wind and rainfall conditions were collected.

In 2009 several field campaigns took place in order to observe the hydrodynamics at the outfall, to measure the velocities of the flow and the water quality parameters in order to validate the model. The measurements were performed with the support of a Bellingardo 550 motorboat utilizing a Geo-nav 6sun GPS system, a Navman 4431 ultrasonic transducer and an YSI556 multi-parameter probe. Morphologic, hydraulic and water quality measurements were executed into the transition estuary of the harbour canal and near the mouth. The dispersion area and profile distribution of freshwater outgoing from the harbour mouth and discharged in the coastal area was investigated and monitored. Experiments were carried out on June 2009 and September 2009. The surface currents were observed with the aim of drifters properly designed to follow the surface pollution and oil (Archetti, 2009). The drifters (Fig. 5) were equipped with a GPS to acquire the geographical position every 5 minutes and an IRIDIUM satellite system was used to send data to a server.

The first experiment was carried out on June 18, 2009. The wave conditions were measured by the wave buoy located 5 nautical miles off the shore of Cesenatico (details on the wave position and data are available at http://www.arpa.emr.it/sim/?mare/boa). The significant

vertical mixing limited to the surface layer.

.

**5.1 Experiment I: June 18, 2009.** 

Fig. 5. Lagrangian drifter in the sea during the experiment.

wave height HS was lower than 0.3 m for the whole day. The measured sea water level and wave conditions on the day of the experiment are plotted in Fig. 6A. The weather conditions were very mild, without wind and with ascending tide, so we had the opportunity to monitor a condition driven only by the tidal excursion. Figure 6B shows the swl during the experiment and the contemporary velocity and direction of the drifters launched 1 km offshore from the Cesenatico harbour canal.

Fig. 6. A) Measured swl (top panel), significant wave height (HS), direction and period (TP). B), drifters' velocity (top panel), direction (central panel) and contemporary swl (bottom panel).

Clusters of three drifters were launched simultaneously at the offshore boundary. The launch position of the drifters is the offshore location in Fig. 7A. The first cluster was launched at about 9:00 a.m. just offshore from the harbour breakwaters, at a distance of 1.2 km from the beach, the second cluster was launched one hour later offshore from the northern beach and the last cluster was launched at 11:00 am offshore from the southern beach. The velocity and direction of the drifters during the experiment is plotted in Fig. 6B. The mean drifter velocity during the experiments was 0.18 m/s, with a direction perpendicular to the beach.

Freshwater Dispersion Plume in the Sea: Dynamic Description and Case Study 141

A

B Fig. 8. A) Measured swl (top panel), significant wave height (HS), direction and period (TP). B) Drifters' velocity (top panel), direction (central panel) and contemporary swl (bottom

A

panel).

Fig. 7. A) Satellite view of the study area and pattern of two drifters launched on June 18, 2009. B). Field for experiment I of surface currents.

The observed condition was simulated by the model; the hydrodynamic was driven only by sea water tidal oscillation at the offshore boundary condition (condition in Fig. 6A ). The resulting surface current field during the experiment condition is shown in Fig. 7B, the current is perpendicular to the shoreline.

The field velocity appears comparable to the drifters' paths, both in direction and magnitude, so the model looks well calibrated.

#### **5.2 Experiment II September 1, 2009**

During the experiment carried out on September 1, 2009, the drifters were launched in the water in a plume of sewage water disposal from the canal of Cesenatico harbour. Two drifters were deployed in the plume centre and two at the plume front. The two drifters deployed at the plume front followed the plume front evolution during the experiment lasting 4 hours. Wind speed was approx. 30 m/s, significant wave height 0.5 m (Fig. 8A) and the tide descending. The plume and the drifters moved in the wind direction at an average speed of 0.2 m/s (Fig. 8B).

A

B Fig. 7. A) Satellite view of the study area and pattern of two drifters launched on June 18,

200 400 600 800 1000 1200 1400 1600

[m]

The observed condition was simulated by the model; the hydrodynamic was driven only by sea water tidal oscillation at the offshore boundary condition (condition in Fig. 6A ). The resulting surface current field during the experiment condition is shown in Fig. 7B, the

The field velocity appears comparable to the drifters' paths, both in direction and

During the experiment carried out on September 1, 2009, the drifters were launched in the water in a plume of sewage water disposal from the canal of Cesenatico harbour. Two drifters were deployed in the plume centre and two at the plume front. The two drifters deployed at the plume front followed the plume front evolution during the experiment lasting 4 hours. Wind speed was approx. 30 m/s, significant wave height 0.5 m (Fig. 8A) and the tide descending. The plume and the drifters moved in the wind direction at an average

2009. B). Field for experiment I of surface currents.

current is perpendicular to the shoreline.

[m]

**5.2 Experiment II September 1, 2009** 

speed of 0.2 m/s (Fig. 8B).

magnitude, so the model looks well calibrated.

Fig. 8. A) Measured swl (top panel), significant wave height (HS), direction and period (TP). B) Drifters' velocity (top panel), direction (central panel) and contemporary swl (bottom panel).

Freshwater Dispersion Plume in the Sea: Dynamic Description and Case Study 143

During the experiments the presence of biological aggregates and foams was observed on the sea surface interested by the plume (Fig. 11). The presence of biological traces in sea areas interested by freshwater dispersion is a well known phenomenon. In a few cases bacterial and dead algae aggregate come directly from internal channels where variation in water depth provides alternance of photosynthetic and bacterial activity. Here, high aerobic biomass levels are produced by bacterial synthesis sustained by the production of photosynthetic oxygen of high growing algae populations. When oxygen, dissolved during light hours, cannot supply nightly bacteria/algal demand, the water column is interested by the presence of many species of died organic substances with the associated settling and floating phenomena. Production of biological foams can occur also when variations in salinity concentrations increase the mortality of a phytoplankton population growth in a low salinity environment. In these cases, foam presence is often registered in the last part of the harbour canal, near the sea mouth, and upon the plume boundary of the sea outfalled

Two vertical profiles of temperature (Fig. 12A), dissolved oxygen, pH, (Fig. 12B) redox potential and salinity concentrations (Fig. 12A) were registered and analysed "on site" in order to check the main plume direction. Fixed investigated points are N1 and S1 focused as representing the north and south near the sea mouth area (see reference map in Fig. 2). Parameters are traced with reference to profile P6 at fixed points located on the east boundary in front of the harbour canal and chosen as indicators of offshore sea conditions. No appreciable variations on salinity vertical distribution are registered in the south zone, where measured values appear very similar in S1 (south near mouth) and P6 (offshore sea). On the contrary, N1 vertical profile presents a salinity distribution which reveals the arrival in the surface layers of volumes coming from the mouth section enriched by internal freshwater. A difference of 2 g/l between bottom and surface layers with thermocline from depth of 60 to 120 cm is registered. Similarly, temperature does not show vertical variations in the south zone, even if media values appear lower in coastal rather than offshore sea water (26.5 °C) according with the cooling effects produced in September by internal water volumes. This is confirmed by the N1 temperature profile which presents lower values in surface layers (25.6°C) than in the underlying thermocline (26.4 °C) but inversion does not interrupt stratification which is maintained by variation in density. Similar temperature values in N1 and S1 points are registered within the thermocline thickness. At thermocline depths a temperature decrease is appreciable due to the colder masses stored at the bottom

N1, N2, N3 points, interested by the dispersion plume, show a pH vertical profile similar to temperature profile. Low pH values usually indicate biological organic substance degradation or nitrification phenomena typically active in waters of internal channels receiving wastewater. In N1 near the mouth point, higher values are confined in a 1 metre thickness layer, sited at a 1 metre depth. On this layer, lower pH values confirm the

Fig. 13 and Fig. 14 show the sequence of profiles obtained following the plume trajectory starting from P1 (internal point corresponding to the slipway) towards to N5 external point placed on the north boundary investigation area. As expected, freshwater volumes are progressively mixed with external high salinity volumes proceeding from internal to external sections. Vertical profiles of salinity behaviour at P1, P2, P3 internal points show that freshwater plume interests a 2 metre depth surface layer. At the last internal section (Gambero rosso), turbulence realizes a linear decrease on salinity concentration from 34 g/l

presence of a plume conditioned by freshwater also indicated by lower temperature.

plume.

of the harbour canal.

Fig. 9. A) Satellite view of the study area and pattern of three drifters launched on September 1, 2009. B). Surface currents' field for experiment II.

Differently from the previous examined condition, we observe here that the drifters' paths are north deviated by the action of the wind on the surface layer with higher velocity (Fig. 9A). The reorientation of the trajectory increases when the drifters approach the coast. Similar behaviour is observed in the hydrodynamic simulation results (Fig. 9B).

The observed and simulated effect is the result of the composition of the marine current driven by tidal oscillation, together with surface wind effect. The described condition is typical in summer in the final hours of the morning.

A model validation was also carried out by comparing simulated and observed salinity vertical profiles into the plume at section N3 during experiment II. The comparison (Fig. 10) shows a good agreement between observed and simulated values also in the vertical profiles. A more extensive comparison of vertical profiles with other parameters and at other sections will be performed in the future.

Fig. 10. Vertical salinity behaviour: observed in point N3 (red) and simulated by the model (blu).

B

Differently from the previous examined condition, we observe here that the drifters' paths are north deviated by the action of the wind on the surface layer with higher velocity (Fig. 9A). The reorientation of the trajectory increases when the drifters approach the coast.

The observed and simulated effect is the result of the composition of the marine current driven by tidal oscillation, together with surface wind effect. The described condition is

A model validation was also carried out by comparing simulated and observed salinity vertical profiles into the plume at section N3 during experiment II. The comparison (Fig. 10) shows a good agreement between observed and simulated values also in the vertical profiles. A more extensive comparison of vertical profiles with other parameters and at

salinity [g/kg] Fig. 10. Vertical salinity behaviour: observed in point N3 (red) and simulated by the model

Fig. 9. A) Satellite view of the study area and pattern of three drifters launched on

Similar behaviour is observed in the hydrodynamic simulation results (Fig. 9B).

September 1, 2009. B). Surface currents' field for experiment II.

typical in summer in the final hours of the morning.

other sections will be performed in the future.

(blu).

During the experiments the presence of biological aggregates and foams was observed on the sea surface interested by the plume (Fig. 11). The presence of biological traces in sea areas interested by freshwater dispersion is a well known phenomenon. In a few cases bacterial and dead algae aggregate come directly from internal channels where variation in water depth provides alternance of photosynthetic and bacterial activity. Here, high aerobic biomass levels are produced by bacterial synthesis sustained by the production of photosynthetic oxygen of high growing algae populations. When oxygen, dissolved during light hours, cannot supply nightly bacteria/algal demand, the water column is interested by the presence of many species of died organic substances with the associated settling and floating phenomena. Production of biological foams can occur also when variations in salinity concentrations increase the mortality of a phytoplankton population growth in a low salinity environment. In these cases, foam presence is often registered in the last part of the harbour canal, near the sea mouth, and upon the plume boundary of the sea outfalled plume.

Two vertical profiles of temperature (Fig. 12A), dissolved oxygen, pH, (Fig. 12B) redox potential and salinity concentrations (Fig. 12A) were registered and analysed "on site" in order to check the main plume direction. Fixed investigated points are N1 and S1 focused as representing the north and south near the sea mouth area (see reference map in Fig. 2). Parameters are traced with reference to profile P6 at fixed points located on the east boundary in front of the harbour canal and chosen as indicators of offshore sea conditions.

No appreciable variations on salinity vertical distribution are registered in the south zone, where measured values appear very similar in S1 (south near mouth) and P6 (offshore sea). On the contrary, N1 vertical profile presents a salinity distribution which reveals the arrival in the surface layers of volumes coming from the mouth section enriched by internal freshwater. A difference of 2 g/l between bottom and surface layers with thermocline from depth of 60 to 120 cm is registered. Similarly, temperature does not show vertical variations in the south zone, even if media values appear lower in coastal rather than offshore sea water (26.5 °C) according with the cooling effects produced in September by internal water volumes. This is confirmed by the N1 temperature profile which presents lower values in surface layers (25.6°C) than in the underlying thermocline (26.4 °C) but inversion does not interrupt stratification which is maintained by variation in density. Similar temperature values in N1 and S1 points are registered within the thermocline thickness. At thermocline depths a temperature decrease is appreciable due to the colder masses stored at the bottom of the harbour canal.

N1, N2, N3 points, interested by the dispersion plume, show a pH vertical profile similar to temperature profile. Low pH values usually indicate biological organic substance degradation or nitrification phenomena typically active in waters of internal channels receiving wastewater. In N1 near the mouth point, higher values are confined in a 1 metre thickness layer, sited at a 1 metre depth. On this layer, lower pH values confirm the presence of a plume conditioned by freshwater also indicated by lower temperature.

Fig. 13 and Fig. 14 show the sequence of profiles obtained following the plume trajectory starting from P1 (internal point corresponding to the slipway) towards to N5 external point placed on the north boundary investigation area. As expected, freshwater volumes are progressively mixed with external high salinity volumes proceeding from internal to external sections. Vertical profiles of salinity behaviour at P1, P2, P3 internal points show that freshwater plume interests a 2 metre depth surface layer. At the last internal section (Gambero rosso), turbulence realizes a linear decrease on salinity concentration from 34 g/l

Freshwater Dispersion Plume in the Sea: Dynamic Description and Case Study 145

A

**NORTH vs SOUTH pH PROFILES**

B Fig. 12. A) Thermoaline and B) pH profiles at the beginning of the experiment at sections S1,

0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 depth (cm)

S1 N1 P6 N3

N1, N3, P6 (see Fig.2).

7,2

7,3

7,4

pH ( )

7,5

7,6

7,7

at 2 m depth to 31 g/l at the surface. This layer overflows upon an almost static high salinity volume placed at the bottom channel. Both P4 and N1 external profiles indicate clear stratification conditions with a 60 cm floating layer. Here, wastewater presence is appreciable and thermocline is located into the underlying 60 cm. Measured salinity surface values together with behaviour of vertical profiles allows the identification of an area interested by plume dispersion limited to a northerly direction by N3 fixed investigation point. Similar profiles at points N4 and N5 reveal that in experiment tidal and currents conditions are typical of offshore sea water volumes.

Fig. 11. View of the floating biological foams observed on the north plume boundary during the September 1, 2009 experiments. Photo taken from the N3 position (see Fig. 2) beach oriented.

at 2 m depth to 31 g/l at the surface. This layer overflows upon an almost static high salinity volume placed at the bottom channel. Both P4 and N1 external profiles indicate clear stratification conditions with a 60 cm floating layer. Here, wastewater presence is appreciable and thermocline is located into the underlying 60 cm. Measured salinity surface values together with behaviour of vertical profiles allows the identification of an area interested by plume dispersion limited to a northerly direction by N3 fixed investigation point. Similar profiles at points N4 and N5 reveal that in experiment tidal and currents

Fig. 11. View of the floating biological foams observed on the north plume boundary during the September 1, 2009 experiments. Photo taken from the N3 position (see Fig. 2) beach

oriented.

conditions are typical of offshore sea water volumes.

**NORTH vs SOUTH pH PROFILES**

Fig. 12. A) Thermoaline and B) pH profiles at the beginning of the experiment at sections S1, N1, N3, P6 (see Fig.2).

Freshwater Dispersion Plume in the Sea: Dynamic Description and Case Study 147

Fig. 14. Vertical profiles of temperature measured at the profile points during experiment

Fig. 15. Vertical profiles of dissolved oxygen at the profile points during experiment

conducted on 1 September, 2009.

conducted on 1 September, 2009.

The sequence of temperature profiles (Fig. 14) reveals very similar vertical trends and values among all profile sections inside the harbour canal (sections P1, P2 and P3). Perhaps a small effect of the external sea water's warmer mass could be noted in the deeper layers at P3 section sited in the proximity of the mouth. Excluding a 40cm sea bottom layer, all points' indicators of dispersion plume area present temperature values lower at surface (N1). As just reported in Fig. 12's comments on comparison of N1 and S1 thermoaline profiles, this initial thermal inversion which does not yet allow a stratification break, confirms salinity indications about plume areal extension. N5 profile, located at the northern boundary investigation area and not interested by colder freshwater coming from the internal basin, maintains a classic summer temperature profile for Adriatic coastal sea. In this case we observe a 26.4 °C constant temperature in a 120 cm depth surface layer, a thermocline to a depth of 240 cm and another 1 metre bottom layer with a constant temperature of 25.2 °C.

Fig. 13. Vertical profiles of salinity measured at the profile points during the experiment conducted on 1 September, 2009.

The sequence of temperature profiles (Fig. 14) reveals very similar vertical trends and values among all profile sections inside the harbour canal (sections P1, P2 and P3). Perhaps a small effect of the external sea water's warmer mass could be noted in the deeper layers at P3 section sited in the proximity of the mouth. Excluding a 40cm sea bottom layer, all points' indicators of dispersion plume area present temperature values lower at surface (N1). As just reported in Fig. 12's comments on comparison of N1 and S1 thermoaline profiles, this initial thermal inversion which does not yet allow a stratification break, confirms salinity indications about plume areal extension. N5 profile, located at the northern boundary investigation area and not interested by colder freshwater coming from the internal basin, maintains a classic summer temperature profile for Adriatic coastal sea. In this case we observe a 26.4 °C constant temperature in a 120 cm depth surface layer, a thermocline to a depth of 240 cm and another 1 metre bottom layer with a constant temperature of 25.2 °C.

Fig. 13. Vertical profiles of salinity measured at the profile points during the experiment

conducted on 1 September, 2009.

Fig. 14. Vertical profiles of temperature measured at the profile points during experiment conducted on 1 September, 2009.

Fig. 15. Vertical profiles of dissolved oxygen at the profile points during experiment conducted on 1 September, 2009.

Freshwater Dispersion Plume in the Sea: Dynamic Description and Case Study 149

A freshwater dispersion plume in the sea has been described in depth in the present paper with the aim of producing a 3D numerical model and with the validation of two field campaigns carried out in different conditions. The investigated area concerns the coastal zone near Cesenatico (Adriatic Sea, Italy). The fresh water is dispersed by the canal harbour

The model shows good performance in the application here presented, which is characterised by the presence of complex sea structures, requiring a very detailed and small

Field data were acquired during two field campaigns and are of different typology: surface lagrangian paths, acquired by innovative properly designed drifters (in both campaigns); vertical profiles of temperature and salinity and dissolved oxygen acquired by a multiparameter probe in properly defined fixed points (in the second campaign). During the first campaign the hydrodynamic was driven only by the tidal oscillation and during the second also by surface wind, the tested conditions were therefore different and interesting

Comparison between model results and measurements are good for the surface hydrodynamic description and for the areal and vertical distribution of concentration, in particular, the resulting salinity values compared with experimental data have shown a

During the second experiment the presence of biological aggregates and foams was observed on the sea surface interested by the plume. The presence of biological traces in sea

Vertical measurement of thermoaline parameters shows appreciable variations on salinity vertical distribution in the southern zone, where measured values appear very similar in the south near mouth and offshore sea. On the contrary, at the northern zone the vertical profiles present a salinitydistribution which reveals the arrival in the surface layers of volumes coming from the mouth section enriched by internal freshwater. A difference of 2 g/l between bottom and surface layers with thermocline from depth of 60 to 120 cm is registered. Similar behaviour was observed for temperature. In fact in the north the temperature profile presents lower values in surface layers (25.6°C) than in the underlying thermocline (26.4 °C), but inversion does not interrupt stratification which is maintained by variation in density. At thermocline depths a temperature decrease is appreciable due to the

The points, interested by the dispersion plume, showed a pH vertical profile similar to temperature profile. Low pH values usually indicate biological organic substance degradation or nitrification phenomena typically active in waters of internal channels receiving wastewater. In N1 near the mouth point, higher values are confined in a 1 metre thickness layer, sited at a 1 metre depth. On this layer lower pH values confirm the presence

The methodology proposed in this paper appears to be useful and accurate enough to

The results here presented are original and have allowed a general comprehension of the

The model now validated can in the future be applied to investigate the dispersion in other

of a plume conditioned by freshwater also indicated by a lower temperature.

simulate the dynamics of the freshwater dispersion at the investigated scale.

thermoaline and hydrodynamic assessment of the dispersion area.

meteo climatic conditions, tides and other canal mouth geometries.

areas interested by freshwater dispersion is a well known phenomenon.

colder masses stored at the bottom of the harbour canal.

**6. Conclusions** 

mouth into the open sea.

mesh dimension in the geometry description.

for understanding the complex dynamics.

surprisingly good agreement.

As expected, oxygen values averaged at each section (Fig. 15) increase, proceeding from internal to external points. At P1 and P2 profiles, photosynthesis produces maximum values in a 60 cm surface layer. At the P3 point (internal but near the mouth), a strong influence of external sea water on bottom layers is confirmed, which shows the same oxygen value, while at surface layers values are typical of internal waters. No information about plume dispersion could be obtained at external points where oxygen distribution is characterised by classic coastal sea profiles with oxygen decreasing values in the direction of deeper layers where photosynthesis is low and bacterial consumption increases.

Results of simulated salinity concentration (Fig. 16), similar to those presented in Fig. 4B, indicate a northerly oriented freshwater dispersion, different from the case analysed in Fig. 4B, which presents in the first phases a less oriented dispersion plume and during the following times (hour 15 – 18) a prevailing orientation to the southern coastal zone. In the actual case, the plume is west bounded by the continuous breakwaters, this means that the geometry is well reproduced in the model, and is dispersed to the north, for the effect of the wind, which was negligible in the previous examined condition.

Fig. 16. Simulation of the freshwater plume dispersion during experiment II.
