COVID-19 Pandemic Consequences on Coastal Water Quality Using WST Sentinel-3 Data: Case of Tangier, Morocco

: The west coast of Tangier, in northern Morocco, has been a ﬀ ected by industrial wastewater discharge that reaches the ocean through the Boukhalef river. Therefore, the Jbila and Sidikacem beaches near to the Boukhalef river mouth have been classiﬁed as polluted for many years. With the aim of determining the COVID-19 pandemic consequences on the Tangier coastal environment, a linear model using Sentinel 3 water surface temperature (WST) has been tested in several locations. Data from April 2019 and April 2020, before and during the COVID-19 pandemic related emergency status in Morocco, were compared. The results from April 2019 showed high WST values and consequently, the poorest water quality in the sites closest to the Boukhalef river mouth. On the other hand, the results from April 2020 showed normal WST values and high water quality in the same study area. These results illustrate the usefulness of Sentinel 3 WST for the estimation of bathing water quality on the west coast of Tangier. The study shows the positive impact of the COVID-19 pandemic consequences on the coastal environment quality in the study area and indicates the importance of decreasing the industrial discharge on the west coast of Tangier. The same methodology could be used in decision-making processes and to reduce cost, time and human resources for coastal monitoring systems. We demonstrate the potential of using the Sentinel 3 data for coastal waters monitoring, as well as the need for stricter controls of pollutant discharges into the world’s rivers.


Introduction
Clean coastal water is an essential requirement for swimmers health and the habitats of aquatic organisms [1]. Coastal water pollution is one of the serious environmental health risks that is affecting most of the world's coastal environments [2,3]. Therefore, monitoring and improving coastal water quality is becoming an inevitable necessity [4][5][6][7][8][9]. Escherichia coli (E. coli) is one of the commonly used bacteria for indicating the fecal contamination in rivers and coastal waters [10][11][12]. The E. coli with minimum delay. They require minimal post-processing as the WST is one of the directly available products. This makes them especially useful for near real-time monitoring.
The contribution of the present work is twofold. First, it illustrates the level of bathing water quality improvement during the COVID-19 lockdown and relates the pollution level of the coastal waters to the rate of industrial activity. Secondly, it demonstrates the possible use of Sentinel-3 WST images as an early warning system of pollution of coastal waters in the area around the Boukhalef estuary and similar areas worldwide. Indeed, the analysis shows a significant change in WST and consequently a marked improvement in the E. coli concentration estimate in the western coastal waters of Tangier during the COVID-19 lockdown period.

Site Description
The study area is situated in the north-western part of Morocco ( Figure 1) on the western coast of Tangier. It is an area of intense industrial activity [6]-about 43% of the automotive industry in Morocco is concentrated in Tangier [29]. The industrial and other human activities, including urbanization and tourism, are concentrated on the Atlantic coast of Tangier [30,31]. All these activities are the main cause of the deterioration of the natural environment in coastal areas [32] and influence bathing water quality [4]. Er-Raioui et al., (2012) [33], reported that more than 232 tons/day (t/d) of suspended solids are discharged in the Tangier and Tetouan coast and that these marine areas are the most affected by human activities in the Moroccan Mediterranean coastline. Previous studies were carried out in the west coast of Tangier [4][5][6][7]34] to assess its bathing water quality, especially considering the industrial growth that Tangier has experienced in the last decades. The reports concluded that the human activities concentrated along the west coast of Tangier have a potential impact on bathing waters, especially at Jbila and Sidikacem beaches. A variety of pollutants could be associated with urban runoff and often containing microbial contaminants which present a risk to human health [35][36][37].
Water 2020, 12, x FOR PEER REVIEW 3 of 18 day and with minimum delay. They require minimal post-processing as the WST is one of the directly available products. This makes them especially useful for near real-time monitoring. The contribution of the present work is twofold. First, it illustrates the level of bathing water quality improvement during the COVID-19 lockdown and relates the pollution level of the coastal waters to the rate of industrial activity. Secondly, it demonstrates the possible use of Sentinel-3 WST images as an early warning system of pollution of coastal waters in the area around the Boukhalef estuary and similar areas worldwide. Indeed, the analysis shows a significant change in WST and consequently a marked improvement in the E. coli concentration estimate in the western coastal waters of Tangier during the COVID-19 lockdown period.

Site Description
The study area is situated in the north-western part of Morocco ( Figure 1) on the western coast of Tangier. It is an area of intense industrial activity [6]-about 43% of the automotive industry in Morocco is concentrated in Tangier [29]. The industrial and other human activities, including urbanization and tourism, are concentrated on the Atlantic coast of Tangier [30,31]. All these activities are the main cause of the deterioration of the natural environment in coastal areas [32]and influence bathing water quality [4]. Er-Raioui et al., (2012) [33], reported that more than 232 tons/day (t/d) of suspended solids are discharged in the Tangier and Tetouan coast and that these marine areas are the most affected by human activities in the Moroccan Mediterranean coastline. Previous studies were carried out in the west coast of Tangier [4][5][6][7]34]to assess its bathing water quality, especially considering the industrial growth that Tangier has experienced in the last decades. The reports concluded that the human activities concentrated along the west coast of Tangier have a potential impact on bathing waters, especially at Jbila and Sidikacem beaches. A variety of pollutants could be associated with urban runoff and often containing microbial contaminants which present a risk to human health [35][36][37]. The main drivers of pollution in the area are two large industrial zones, Tangier Free Zone (TFZ) and Gzenaya Zone (GZ), which extend over a surface of 48.12 km 2 . The area includes the Boukhalef river and the Boukhalef wastewater treatment plant [6]. The TFZ and GZ are known as the most important industrial zones in the northern part of Morocco [4,7,38]. They represent the core of industrial activities of Tangier city and contain more than 715 industrial units and companies [6] including: agri-food industries (AFI), chemical, para-chemical industries (CPI), textile and leather industries (TLI), mechanical, metallurgical, and electronics industries (MMEI) and other services The main drivers of pollution in the area are two large industrial zones, Tangier Free Zone (TFZ) and Gzenaya Zone (GZ), which extend over a surface of 48.12 km 2 . The area includes the Boukhalef river and the Boukhalef wastewater treatment plant [6]. The TFZ and GZ are known as the most important industrial zones in the northern part of Morocco [4,7,38]. They represent the core of industrial activities of Tangier city and contain more than 715 industrial units and companies [6] including: agri-food industries (AFI), chemical, para-chemical industries (CPI), textile and leather industries Water 2020, 12, 2638 4 of 18 (TLI), mechanical, metallurgical, and electronics industries (MMEI) and other services industries (OSI). The untreated and treated wastewaters released from these industrial activities, which are often characterized by a high concentration of pollutants, are discharged directly into the Boukhalef river and eventually reach the seacoast at Sidikacem beach ( Figure 2) [39]. Indeed, Jbila and Sidikacem beaches, located near the Boukhalef River mouth, are known to be chronically contaminated [4,6,8,33]. In addition, those two beaches are the most popular tourist destinations for swimmers with a frequency of up to 8000 individuals per day during the summer season [6].
Water 2020, 12, x FOR PEER REVIEW 4 of 18 industries (OSI). The untreated and treated wastewaters released from these industrial activities, which are often characterized by a high concentration of pollutants, are discharged directly into the Boukhalef river and eventually reach the seacoast at Sidikacem beach ( Figure 2) [39]. Indeed, Jbila and Sidikacem beaches, located near the Boukhalef River mouth, are known to be chronically contaminated [4,6,8,33]. In addition, those two beaches are the most popular tourist destinations for swimmers with a frequency of up to 8000 individuals per day during the summer season [6].

Study Procedure
In this study, the western Tangier water quality was analyzed using Sentinel 3 images as shown in Figure 3 with details in the following sections.
The WST data for April 2019 and 2020 were extracted from Sentinel 3 images, the first year representing a period of normal industrial production activity in both industrial zones, the Tangier Free Zone and Gzenaya Zone (Figure 2), and the second year representing a period of halt in industrial production due to the Covid-19 pandemic emergency. The linear coastal water quality model developed by   [4], was used to estimate the E. coli concentration and to estimate the Tangier western coastal waters quality ( Figure 3).

Study Procedure
In this study, the western Tangier water quality was analyzed using Sentinel 3 images as shown in Figure 3 with details in the following sections.
The WST data for April 2019 and 2020 were extracted from Sentinel 3 images, the first year representing a period of normal industrial production activity in both industrial zones, the Tangier Free Zone and Gzenaya Zone (Figure 2), and the second year representing a period of halt in industrial production due to the Covid-19 pandemic emergency. The linear coastal water quality model developed by Cherif et al., (2019) [4], was used to estimate the E. coli concentration and to estimate the Tangier western coastal waters quality ( Figure 3).
Water 2020, 12, x FOR PEER REVIEW 5 of 18 Figure 3. Flowchart of the Tangier western coastal water quality investigation using Sentinel 3 water surface temperature (WST) data.

Water Surface Temperature Data
The WST measurements are carried out based on Level-2 WST products derived from the SLSTR (Sea and Land Surface Temperature Radiometer) instrument. The latter has a very wide swath of 1400 km and covers the entire earth surface in about 2 days. It can provide a map of ocean temperatures with extreme accuracy up to 0.1 degrees. SLSTR records the same area from two different angles, which enables (by combining the two images) the correction of atmospheric disturbance effects in order to obtain very precise temperature measurements. SLSTR has 9 channels in visible and infrared (3 in visible and near-infrared, 3 in SW infrared, and 3 in MW infrared). The set of spectral bands covers the range from 0.556 µm to 12.5 µm. More details about the WST products can be found on the official website: https://sentinel.esa.int/web/sentinel/user-guides/sentinel-3-slstr/product-types/level-2-wst. It is also worth mentioning that the SL_2_WST products can be obtained from the Group for High-Resolution Sea Surface Temperature (GHRSST) https://www.ghrsst.org/.

Water Surface Temperature Data
The WST measurements are carried out based on Level-2 WST products derived from the SLSTR (Sea and Land Surface Temperature Radiometer) instrument. The latter has a very wide swath of 1400 km and covers the entire earth surface in about 2 days. It can provide a map of ocean temperatures with extreme accuracy up to 0.1 degrees. SLSTR records the same area from two different angles, which enables (by combining the two images) the correction of atmospheric disturbance effects in order to obtain very precise temperature measurements. SLSTR has 9 channels in visible and infrared (3 in visible and near-infrared, 3 in SW infrared, and 3 in MW infrared). The set of spectral bands covers the range from 0.556 µm to 12.5 µm. More details about the WST products can be found on the official website: https://sentinel.esa.int/web/sentinel/user-guides/sentinel-3-slstr/product-types/level-2-wst. It is also worth mentioning that the SL_2_WST products can be obtained from the Group for High-Resolution Sea Surface Temperature (GHRSST) https://www.ghrsst.org/.
Once the suitable images were downloaded, the WST values corresponding to our points of interest were extracted. Figure 4 shows the following steps using SNAP (Sentinel Application Platform) software developed by the European Spatial Agency (ESA) and is freely available from their official website: http://step.esa.int/main/download/snap-download/.  Year Images ID Once the suitable images were downloaded, the WST values corresponding to our points of interest were extracted. Figure 4 shows the following steps using SNAP (Sentinel Application Platform) software developed by the European Spatial Agency (ESA) and is freely available from their official website: http://step.esa.int/main/download/snap-download/.

Visual Analysis
The WST data were analyzed using the Matlab environment. A sample image is presented in Figure 5. It depicts the temperature values represented by different colors, ranging from blue for low temperature to dark red for high temperature values. The WST data were analyzed using the Matlab environment. A sample image is presented in Figure 5. It depicts the temperature values represented by different colors, ranging from blue for low temperature to dark red for high temperature values.

Coastal Water Quality Estimation
In order to determine the Tangier coastal water quality, the equation developed by Cherif et al., (2019) [4] was used. It is a linear model based on the correlation between bacteriological in situ measurements and thermal data from Landsat 8 satellite: Following the equation below (1), we first determined the value of variable X ( * ) and then calculated the estimated values of E. coli concentration (Y), including the root mean square deviation (RMSD). Quality classes were assigned following the Moroccan quality standards according to the Moroccan norm NM 03.7.200 [40] (see Table 2). Y = 0.37 X + 2.4; RMSD = 0.5 (1) where: where WST represents the water surface temperature and Ta represents air temperature. The air temperature (Ta) was obtained from the Tangier Ibn Batouta airport weather station (https://www.wunderground.com/history/daily/ma/hjarennhal/GMTT/date/2019-4-2).

Coastal Water Quality Estimation
In order to determine the Tangier coastal water quality, the equation developed by Cherif et al., (2019) [4] was used. It is a linear model based on the correlation between bacteriological in situ measurements and thermal data from Landsat 8 satellite: Following the equation below (1), we first determined the value of variable X ( * ) and then calculated the estimated values of E. coli concentration (Y), including the root mean square deviation (RMSD). Quality classes were assigned following the Moroccan quality standards according to the Moroccan norm NM 03.7.200 [40] (see Table 2).

Results and Discussion
Given that RS WST data have already been used in an exploratory study to determine the coastal water quality in the north of Morocco [4], we tested the possibility of the developed linear equation (between E. coli and WST data) providing significant results for examining the impact of reduced industrial activities on the coastal water quality. The COVID-19 pandemic immobilized the TFZ and GZ activities, which presented a convenient opportunity to study the western Tangier coastal water quality in low wastewater outflow period. Since it was not possible to take in situ measurements, the Sentinel 3 data seemed to be an adequate tool to assess the possible change of bathing water quality in the studied area. The Sentinel 3 data have already proved useful in monitoring high levels of pollution in coastal areas, and presents a new tool for surveying the ocean [41,42], as well as the runoff of many river mouths [43].
The preliminary results of the western Tangier coastal water quality obtained by Cherif et al., (2019) [4] that were based on Landsat 8 imagery, namely, Thermal Infrared Sensor (TIRS) bands were analyzed to monitor a narrow strip of near-shore water where increased levels of pollution had been previously observed. The area of interest included the Boukhalef river mouth and both nearby beaches, Jbila and Sidikacem, as well as non-polluted waters at Sol, Bakacem, Achekar, and Diplomatic Forest beaches, which are quite distant from the river estuary. It was reported in a previous study carried out in this Tangier coastal area [5] that high spatial resolution images, such as those from ESA Sentinel 2 imaging spectrometer, often revealed patches of increased water reflectance values of the Boukhalef river mouth.
An observation example is given in Figure 6, which shows a WST Sentinel 3 acquisition dated on 5 April 2019 and 8 April 2020 (10h00m UTC). A closer inspection quickly reveals areas with very low WST values north and south of the Boukhalef estuary. Between those, an area of increased WST values is usually located and can be clearly seen in the April 2019 image ( Figure 6, April 2019). In April 2020 this warm-water feature is missing and normal WST values were registered at all the sites ( Figure 6, April 2020). Figure 7 shows the WST variation in the Tangier western coastal waters over ten different days in April 2019 and eight different days in April 2020. Generally, the WST was characterized by differentiation and variation of values from the north to the south of the Boukhalef river mouth. In April 2019 the highest values of WST were registered at S4 (880 m from the river mouth), S5 (river mouth) and S6 (755 m from the river mouth), while lower values were recorded at S1 (5300 m from the river mouth), S2 (4600 m from the river mouth), S3 (4000 m from the river mouth) and S7 (6100 m from the river mouth) (Figure 7a). In April 2020 almost all sites registered similar WST values without a significant difference between S5 (river mouth) and other sites (Figure 7b). The April 2019 WST results (Figure 7a) are similar to those from 2017 [4]. These findings confirm that the Boukhalef outflow (S5) has a significant impact in Jbila and Sidikacem beaches (S4 and S6) [7]. A series of coastal water samples along the Atlantic were analyzed by the ministry of environment in 2019, which outlines in the final report the vulnerability of both beaches near the Boukhalef river mouth, especially Jbila [44]. However, the results from April 2020 (Figure 7b) demonstrate a significant change in WST values-especially in locations marked S4, S5, and S6. It is a well-known fact that high-temperature wastewater generated from industrial activities is among the main factors of increased ocean water surface temperature [45]. Thus, the observed decrease in temperature values in the studied stations is due to the reduction of the industrial effluent from the Boukhalef river mouth and the reduction of organic charge concentration in the Boukhalef river waters responsible for heavy bacterial contamination [31,46].  Figure 7 shows the WST variation in the Tangier western coastal waters over ten different days in April 2019 and eight different days in April 2020. Generally, the WST was characterized by differentiation and variation of values from the north to the south of the Boukhalef river mouth. In April 2019 the highest values of WST were registered at S4 (880 m from the river mouth), S5 (river mouth) and S6 (755 m from the river mouth), while lower values were recorded at S1 (5300 m from the river mouth), S2 (4600 m from the river mouth), S3 (4000 m from the river mouth) and S7 (6100 m change in WST values-especially in locations marked S4, S5, and S6. It is a well-known fact that high-temperature wastewater generated from industrial activities is among the main factors of increased ocean water surface temperature [45]. Thus, the observed decrease in temperature values in the studied stations is due to the reduction of the industrial effluent from the Boukhalef river mouth and the reduction of organic charge concentration in the Boukhalef river waters responsible for heavy bacterial contamination [31,46]. The difference in average X value (X = WST−Ta; input variable in Equation (1) is shown in Figure 8. In the left image (average X value in 2019), an area where the WST values are significantly higher than the air temperature is clearly visible-the river mouth (S5) as well as Jbila beach (S4) and Sidikacem beach (S6). While lower or similar WST values to the air temperature can be observed at Sol beach (S1), Bakacem beach (S2), Achekar beach (S3) and Diplomatic Forest beach (S7). The sites S1, S2, S3, S4, S5, S6, and S7 exhibit WST values lower than the air temperature during April 2020 (Figure 7, right).
According to Figure 8, the average difference of WST values and air temperature (X ( * ) ) demonstrates the negative impact of the Boukhalef river waters, which discharge into the ocean (S5) after running along a heavily industrialized area (TFZ and GZ), in the western Tangier coastal water quality. The high-temperature values indicate a high organic pollution concentration in the Boukhalef river runoff and are in good agreement with Dahiya et Kaur, (1999) [47]. The increased near-coastal water temperature of up to 4 °C in 2019 compared to 2020 at the mouth of the Boukhalef river (Figure 8  The difference in average X value (X = WST−Ta; input variable in Equation (1) is shown in Figure 8. In the left image (average X value in 2019), an area where the WST values are significantly higher than the air temperature is clearly visible-the river mouth (S5) as well as Jbila beach (S4) and Sidikacem beach (S6). While lower or similar WST values to the air temperature can be observed at Sol beach (S1), Bakacem beach (S2), Achekar beach (S3) and Diplomatic Forest beach (S7). The sites S1, S2, S3, S4, S5, S6, and S7 exhibit WST values lower than the air temperature during April 2020 (Figure 7, right).
According to Figure 8, the average difference of WST values and air temperature (X ( * ) ) demonstrates the negative impact of the Boukhalef river waters, which discharge into the ocean (S5) after running along a heavily industrialized area (TFZ and GZ), in the western Tangier coastal water quality. The high-temperature values indicate a high organic pollution concentration in the Boukhalef river runoff and are in good agreement with Dahiya et Kaur, (1999) [47]. The increased near-coastal water temperature of up to 4 • C in 2019 compared to 2020 at the mouth of the Boukhalef river (Figure 8) is thus a result of the increased industrial activity/effluent during pre-Covid19 time. These results confirm the investigation results about the source of bathing water pollution in the area found by Cherif et al., (2019) [7]. On the other hand, this figure reflects the positive impact of the suspension of industrial activities on bathing water quality improvement in the investigated area. Figure 9 exhibits the estimated variation of E. coli concentration using the linear model during April 2019 and April 2020 in the west coast of Tangier. The results indicate the presence of high concentrations of E. coli in coastal water sites during April 2019, while lower values of E. coli were recorded during April 2020. According to Schernewski et al., (2012) [48], in coastal and transitional waters, a "sufficient" water quality class allows a maximum of 500 (185) Escherichia coli (intestinal enterococci) colony-forming units (CFU)/100 mL based upon a 95-percentile valuation. For an "excellent" quality class, the highest level, a maximum of 250 (100) E. coli CFU/100 mL is tolerated. The values obtained during April 2020 are almost below 200 E. coli CFU/100 mL. In contrast, the E. coli values in April 2019 were high, confirming that the aforementioned coastal water sites are infected by the industrial wastewater (from TFZ and GZ), which is loaded with a high concentration of nutrient and organic matter responsible for the reproduction of bacteria [49] and demonstrate the existence of an anaerobic environment responsible for the release of bad odors [39,50].   "excellent" quality class, the highest level, a maximum of 250 (100) E. coli CFU/100 mL is tolerated. The values obtained during April 2020 are almost below 200 E. coli CFU/100 mL. In contrast, the E. coli values in April 2019 were high, confirming that the aforementioned coastal water sites are infected by the industrial wastewater (from TFZ and GZ), which is loaded with a high concentration of nutrient and organic matter responsible for the reproduction of bacteria [49]and demonstrate the existence of an anaerobic environment responsible for the release of bad odors [39,50].    [4], and further confirms the negative impact of the TFZ and GZ on the bathing water quality in the area. Locations S1, S2, and S3 were classified as less polluted in most cases (A) and S7 was generally classified as B (Figure 9, April 2019).
The results for April 2020 (when almost 80% of the industrial activity in TFZ and GZ was suspended due to the COVID-19 related lockdown) are shown in Figure 9-all the sites were classified as A (good quality) (Figure 9, April 2020). This indicates the decrease of the TFZ and GZ wastewater discharge into the Boukhalef river and, consequently, the improvement of bathing water quality of the nearest sites to the Boukhalef river mouth (S5), S4 and S6 (Jbila and Sidikacem beaches).  The results for April 2020 (when almost 80% of the industrial activity in TFZ and GZ was suspended due to the COVID-19 related lockdown) are shown in Figure 9-all the sites were classified as A (good quality) (Figure 9, April 2020). This indicates the decrease of the TFZ and GZ wastewater discharge into the Boukhalef river and, consequently, the improvement of bathing water quality of the nearest sites to the Boukhalef river mouth (S5), S4 and S6 (Jbila and Sidikacem beaches).
There is a slight possibility that our results could be influenced by an unusually low temperature or unusually low flow rate of the Boukhalef river. To eliminate this possibility we compared the average air temperature at Ibn Batouta airport weather station (used as a proxy for river water temperature) in April 2019 (Ta = 20 • C) with the average air temperature in April 2020 (Ta = 19.5 • C). We would also like to emphasize that the air temperature (proxy for river temperature) has already been taken into account in our linear model (Equation (1), value X ( * ) ) in order to reduce the sensitivity of results to river water temperature. The average precipitation in the Boukhalef watershed was 66 mm in April 2019 and 67 mm in 2020 (Agence du Bassin Hydraulique du Loukkos, Morocco-private communication). Similarly the weather station in Tangier registered 99.1 mm of rain in April 2019 and 91.7 mm of rain in April 2020 (Agence du Bassin Hydraulique du Loukkos, Morocco-private communication). Based on the precipitation measurements we can conclude that the flow rate was similar in both years. Our results exposed the coastal waters quality status before and during the COVID-19 pandemic using Sentinel 3 data. The use of satellite imagery also enabled us to study the spatial extent of the impact of the industrial zones (TFZ and GZ) on the bathing water quality in one of the most important touristic destinations in northern Morocco. The Sentinel 3 WST offers suitable daily images that help observing the coastal changing conditions. However, it is important to take into account the cloud obstruction, which is sometimes masking our study area. The results obtained confirm the fact that the lockdown caused by COVID-19 was an excellent occasion to assess the industrial activity impacts on the western coastal waters of Tangier. According to Wang and Su, (2020) [51], declining human activities lead to decreasing pollution levels and, consequently, the safety measures taken to fight the pandemic were a benefit to the environment. Several studies have also confirmed the positive impacts of the imposed industrial activities halt on the environment quality [20,[52][53][54][55]. In the most polluted river in India (Vembanad lake), the suspended particular matter concentration was reduced by 15.9% after halting human activities due to COVID-19 [56]. Similar results have been found in a coastal industrial zone in south India which has experienced a significant decrease of pollutants owing to the reduced wastewater discharges from industrial units during the imposed lockdown [57]. Other studies have also proved that the lockdown caused by COVID-19 spread has contributed to the reduction of pollution levels in coastal areas [58][59][60][61]. Hence, these findings in combination with our work present a clear picture of the impact of economic activities on the environment and provide a baseline reference for future studies assessing the possible reduction of pollution levels in the world's coastal waters.

Conclusions
The COVID-19 pandemic and its subsequent control measures led to a significant reduction in industrial activities in the Tangier Free Zone (TFZ) and Gzenaya Zone (GZ) which are significant contributors to the pollution of the Boukhalef river. The latter discharges into the Atlantic ocean close to several popular bathing locations. Due to unprecedented reduction in industrial activity and lack of in-situ data, the COVID-19 lockdown presented a perfect opportunity to use remote sensing and our linear model [4] to estimate changes in coastal waters pollution.
It was somewhat expected that the decline in industrial activity would contribute to the improvement of bathing water quality in the area around the Boukhalef river estuary, but the degree of change is truly remarkable. Not only is the change clearly visible through satellite images but, according to our results, all points of interest could be marked with the A quality class at all observed periods during the COVID-19 related lockdown. One year ago, these same locations exhibited many cases marked with a C, or even D, quality class during the same period.
Through our study, we demonstrated the usefulness of the Sentinel-3 sea surface temperature (SST) images in water quality applications. By using the method introduced by Cherif et al., (2019) [4], the obtained SST and air temperature can be converted directly to E. coli concentrations. The method has been calibrated and tested on the Atlantic coast of Tangier but could be used in any similar location with proper recalibration.
Our work shows that satellite SST measurements represent a viable low-cost solution for bathing water quality monitoring and could be used in combination with field sampling. The demonstrated reliability of satellite-derived sea water temperature as a proxy for E. coli concentration would reduce the required frequency of costly field campaigns. The high frequency of Sentinel 3 passes and high image resolution make it especially suitable for such purposes. Such an approach could also serve as a primary component of a potential early warning system. An optimal system, however, would consist of a regional high-resolution ocean model with SST assimilation and input from a hydrological station near the Boukhalef river mouth, which would provide real-time outflow and water temperature measurements. Such a system would ensure continuous water quality estimates even in periods with poor satellite coverage due to cloudiness.
The results of this study show that the intensity of industrial activity has a strong effect on the bathing water quality in the area surrounding the Boukhalef river estuary. With more accurate data on the rate of activity in each of the industrial units in both study periods, our results could be used for evaluating the effects of future environmental policies and limitations imposed on different industrial sectors on the water quality in the area.