**2.2 Pilot Study**

A numerical experiment was carried out, using 6 months of *E. coli* concentration data in the mollusc flesh and intravalvular liquid, detected in three pilot areas around the Pescara River mouth (**Figure 1**). Since official discharge data from hydrological annals are available until 2010, for the same 6-month period, a hydrological simulation was performed by using the CHyM distributed hydrological model [53–55], forced with observed rainfall data. The model has been extensively used in Abruzzo Region for flood forecast activities [56, 57]. The hydrometeorological conditions preceding each *E. coli* concentration exceedance were investigated, in terms of accumulated rainfall in the coastal area and runoff in the Pescara River mouth. A quick overview of obtained results is here given and discussed; more detailed information is available in Colaiuda et al. [41].

The analysed catchment originates in the inner, northern part of the Abruzzo region, draining an area of about 3147 km2 before flowing into the Adriatic Sea. It is characterised by a very complex orography with altitudes spanning from zero up to almost 3000 m a.s.l. in the range of 150 km. The last ten kilometres along the river path are strongly urbanised, with a relevant solid transport amount, estimated at 106 tons/year, considering only the Pescara city urban area.

Outcomes of hydrometeorological investigations linked to the *E. coli* concentration peaks suggested that *i) E. coli* concentrations appeared to be most linked to the discharge peaks with respect to the precipitation values and *ii) E. coli* peaks

**Figure 1.** *Three pilot areas around the Pescara River mouth.*

*Coastal Water Quality: Hydrometeorological Impact of River Overflow and High-resolution… DOI: http://dx.doi.org/10.5772/intechopen.104524*

exceeding the reported threshold occurred after 2 or 3 days after the Pescara River discharge peak, in most cases.

In more detail, 29 samplings were analysed, and exceeding *E. coli* concentrations were linked to a runoff overflow in 83% of cases and a rainfall event in 50% of cases. As for the mussel farm sampling location, in the open sea at ~5km south-east the Pescara mouth, the 100% of exceeding *E. coli* concentrations were linked to a river overflow, while only the 33% were preceded by rainfall in the coastal area. The case study that occurred on March 8, 2016, revealed a high peak of *E. coli* without any river runoff increase a few days before the event. This case study was then deepened and the hydrometeorological analysis revealed that a huge discharge peak, reaching about 400 m3 /s, affected the Pescara River 7 days before, suggesting a longer river effect on bacterial transport for this case. Moreover, in some cases, a precipitation event over coastal areas and a river discharge increase occurred at the same time and the contribution of the two forcings cannot be discriminated at a first glance. The rainfall effect may also include the presence of sewer overflows (CSOs), direct land-runoff into the estuary, and re-suspension of contaminated sediments within the estuary itself. Finally, increased levels of *E. coli* in bivalves from all monitoring points under high river flow conditions suggest that stormwater runoff is contributing to a significant proportion of *E. coli* accumulation in bivalves [46]. Nevertheless, due to the catchment extension and geographical location, the coastal rainfall does not represent an environmental descriptor indicative of possible faecal contamination related to weather events. The discharge overflow estimation is indeed more representative of the hydrometeorological precursor (**Figure 2**).

A significant association between *E. coli* concentrations and the magnitude of the antecedent discharge peak has been carried out [41]. The Spearman's correlation coefficient rD calculated was 0.69, and the associated *p*-value was low (∼4.5 × 10 5), confirming the correlation hypothesis. The correlation between rainfall maxima and *E. coli* concentrations resulted in a lower correlation coefficient (rR ¼ 0.35) and the associated *p*-value was high (∼0.065), not confirming the correlation hypothesis.

#### **Figure 2.**

*Time series showing E. coli concentrations at P1, P2, and Mussel Farm, the discharge at the mouth of the Pescara River from November 2015 to May 2016.*

Hydrological conditions prior to river flow peaks, such as heavy rainfall, are important in determining the presence of *E. coli* in seawater, but it cannot be ruled out that even low rainfall events could cause significant increases in concentrations when they follow a dry period.

Local regulations for monitoring water and molluscs have been planned regardless of weather conditions, the river flows, or other abiotic factors that can affect the concentration of FIO, and sampling intervals to detect the potential microbial contamination may, therefore, not be representative of variations in these conditions. For this reason, it is reasonable to assume that the data underestimate the strength of the correlations between bacterial concentration, precipitation, and river flow.

To overcome these limits, a holistic approach based on the correlations between data of precipitation (intensity and position) and variation in the river flow discharge is essential. This strategy is useful for predicting times and places of exposure to microbiological contamination. The combined assessment of abiotic factors (physical and chemical), hydrometeorological components and biotic factors also provides holistic information on the health of the ecosystem.
