**3. High-resolution coastal mapping of Chl-a concentration from Sentinel-2 satellite**

As previously introduced, the use of satellite data is reaching constantly wider applicability in the earth monitoring sector. The availability of free satellite data throughout the globe, characterised by a great variety of accessible sensed data types, has sped up this process. In Europe, the raise of programmes, such as Copernicus (European Union's Earth Observation Programme) founded by the European Commission, allowed the birth of new value-adding activities and studies on earth monitoring and services for disaster prevention [58, 59]. The whole programme is composed of seven missions, with a fleet of around ten satellites among future and operational ones. The most important thematic streams of Copernicus services are dedicated to land and marine environments. An example of marine applications is Sentinel-3 satellites that have a push-broom imaging spectrometer called OLCI (Ocean Land Color Instrument). This instrument measures solar radiation reflected by the Earth in 21 spectral bands and has a ground spatial resolution of 300 m [60].

For what concerns the marine environment, as above extensively analysed, its health status is strictly connected to many human activities, especially for coastal waters which represent a vital asset. The possibility to monitor it constantly and rapidly, covering large portions of territory, plays a crucial role and perfectly fits the enhancement introduced using satellite data for monitoring.

The symptoms to be monitored for marine environment's health status evaluation are several and an example is the detection of algae presence. Under certain conditions, indeed, algae can reproduce in an accelerated way, giving birth to what is called an "algal bloom". Some kinds of algal bloom can be toxic, may cause skin rashes or illness in humans as well as can be poisoning for some marine species (e.g., shellfish). A possible parameter to measure algae presence in a water body is chlorophyll-a (Chl-a), the pigment that is used by algae for photosynthesis which constitutes the part that mostly interacts with solar radiation. The concentration of Chl-a contributes to the so-called particulate organic matter present in a water body. On the other hand, water clarity is also connected to all the possible suspended matter that contributes, when in high concentrations, to increase its turbidity [61, 62]. The concentration of all possible suspended particles in a water body is defined as Total Suspended Matter (TSM) concentration. Coastal waters, moreover, operate as a link between land and ocean systems. Rivers physically allow this connection, acting as

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

a conduit for delivering significant amounts of dissolved and particulate materials from terrestrial environments to the coastal ocean, increasing TSM concentrations. In some cases, part of this TSM can be composed of soil particles detached from the coastline and dragged away from waterpower [63].

Several approaches have been followed for years to perform marine monitoring through satellite observations, depending on the specific parameter to be estimated (Sea Surface Height, Wind Speed, Sea Ice, just to mention a few) [64]. For what concerns the detection of water quality, the estimated parameters are connected to its bio-optical properties. At some specific wavelengths, indeed, the suspended particles inside water can interact with radiation incoming from the atmosphere, giving back in return an upwelling radiant flux that has some characteristic responses (it can be absorbed in certain wavelengths more than at others). Optical sensors, therefore, have been widely used to identify the so-called spectral marine inherent optical properties (IOPs, e.g., absorption and scattering) [65]. Usually, the water-leaving signal is quite low (sometimes 1% or less of downwelling irradiance) and requires the sensors to work in a set of narrow, sensitive spectral channels and to remove the atmospheric effects. Those sensors are usually referred to as "ocean color" sensors. The spectral signals received can be used to estimate phytoplankton abundance and other radiatively active constituents.

Usually, the main approaches followed to retrieve IOPs from satellite measurements are two [66]—the first applies atmospheric correction (AC) algorithms to remove the contribution of the atmosphere from the signal received at the top of the atmosphere (TOA) by the sensor and leads to the estimate of the bottom of atmosphere (BOA) reflectance (calculated as the ratio of water-leaving radiance to downwelling irradiance just above the air-sea interface). Different kinds of algorithms can be then applied to AC reflectance values to produce estimates of geophysical properties (e.g., inversion model [67, 68]). The second approach tries to find a direct relationship between the spectral radiance at the top of the atmosphere and IOPs [69, 70]. This one is more immediate and removes the AC step that can sometimes lead to misinterpretation of the atmospheric contribution in presence of optically complex water masses.

However, the resolution of the satellite images used remains the main constraint for accuracy and precision obtained through monitoring developed solutions. It is important to note that coastal areas are also spatially and optically complex and would require more frequent spatial and spectral sampling to enhance the monitoring capability [71].

#### **3.1 Advantages of Sentinel-2 satellite data for coastal water remote sensing**

As said, satellite data can boost the realisation of more effective environment monitoring algorithms. Among the most used satellite data for coastal water studies, we can find several studies that employ high-resolution optical data obtained from the OLI (Operational Land Imager) sensor on board of Landsat-8 satellite and MERIS (Medium Resolution Imaging Spectroradiometer) on the Envisat satellite [72, 73]. Additionally, to the OLI sensor, Landsat 8 satellite payload is also made of Thermal Infrared Sensor (TIRS). These two sensors are characterised by a spatial resolution of 30 meters (visible, NIR, SWIR); 100 meters (thermal); and 15 meters (panchromatic). MERIS, which is a push broom radiometer, reaches a spatial resolution of 300 m at nadir (for full resolution products) and 1200 m for reduced resolution data and its spectral range varies from 390 nm to 1040 nm.

In parallel with the aforementioned missions, and despite being built mainly as a land monitoring mission, also Sentinel-2 satellite has gained popularity for marine applications. Indeed, thanks to its high spatial resolutions together with a high

revisit frequency, Sentinel-2 allowed to overcome several limitations of existing missions. Sentinel-2 is equipped with a MultiSpectral Instrument (MSI) with 13 spectral bands from the visible and near-infrared to the short-wave infrared (from 443 to 2190 nm). The spatial resolution varies from 10 m to 60 m, depending on the spectral band, with a 290 km field of view [74]. The MSI sensor is made of a three-mirror, 150 mm aperture telescope which collects light and focuses it into two separate focal planes—one for visible (VIS) and near-infrared (NIR), and the other for short-wave infrared (SWIR) wavelengths, respectively. Each focal plane is composed of 12 detectors staggered in two rows.

Its revisit frequency is increased with respect to other missions thanks to the simultaneous operations of two identical satellites: Sentinel-2A and Sentinel-2B, launched in 2015 and 2017, respectively. This more frequent data availability offers several benefits—from a higher probability of finding imagery clear of cloud and sun glint; to more effective applicability of change detection algorithms [75]. Moreover, the free availability of its data allowed to facilitate the spread of their usage.
