Context

April 2020, the coronavirus continues to spread around the world like a gust of wind. Hotbeds of contamination are multiplying, cases of disease number in the hundreds of thousands, the world economy is spatially and temporally frozen, humanity is on pause. And while mankind is confined, just about everywhere on the planet, nature is taking back its rights. People no longer go out to work, cars remain in garages, airplanes are landed firmly on the Earth … and cruise ships and private boats remain in port. Indeed, since the beginning of the spread of the virus, there has been a sharp decrease in the number of cruises, and with good reason! On the Diamond Princess ⁽¹⁾, for example, 700 passengers were infected during the cruise, enough to make everyone who heard about this want to stay home.

Shipping companies are starting to offer reduced rates for the end of summer 2020 because for this month of April and the rest of the containment, international tourist activities are completely frozen.This context obviously induces a financial blow for the cities that used to be the destinations of these cruises, such as Marseille in France, Dubrovnik in Croatia or Venice in Italy. However, in the latter, an ecological phenomenon is emerging that is buzzing in the media at the moment. In Venice, fish can be seen in the canals ⁽²⁾ and birds are coming to land in clearer waters than they have been for a very long time.

Simple flu virus or signal from our planet Earth? What is certain is that COVID-19 and its proliferation have a seemingly significant impact on pollution through the collective measures that states enforce in their cities and countries.

At Murmuration, we use satellite measurements of the reflectance (what the satellite sees, the light reflected from the Earth) of the waters to calculate turbidity, and thus correlate pollution and coronavirus advancement. This is what we will develop little by little in this article.

Turbidity

In this second article on the impact of covid-19 on the environment, we will focus on the evolution of water quality in an area with a high level of maritime activity in normal times.

In the last article we demonstrated why NO2 was a good indicator of air quality. This time, the proxy we use is water turbidity. A proxy is a variable that is not significant, but replaces a useful but unobservable or unmeasurable variable (here pollution or water quality). 

Turbidity is the content of clouding matter in a fluid. In other words, the greater the turbidity, the less clear the water and the less you can see through it. This can allow us to measure pollution and its evolution in European hot spots, capitals or even maritime areas.Indeed, we should be aware that turbidity comes mainly from currents caused by passing ships lifting sediments and nutrients from the bottom of the water. These suspended solids contribute to the eutrophication of the water. Eutrophication of aquatic environments is an imbalance in the environment caused by an increase in the concentration of nitrogen and phosphorus in the environment. The level of suspended matter in the water, and therefore turbidity, is an important indicator of water quality and very often explains the presence or absence of fauna in the environments concerned.

Our friends the satellites and researchers in French Guiana

Copernicus is the European Earth observation program, manages by the European Commission. Its satellites belong to the European Union. It is an ambitious and unique programme that looks at our planet and its environment for the benefit of all European citizens and the wider international community. It offers information services based on satellite Earth observation and in situ (non-space) data.

Copernicus has enabled the launch of a group of satellites called “sentinels”. Among all these observers, it is the class 2 of the sentinels that will be of interest here. 

The Sentinel-2 satellites were put into orbit in 2015 and 2017 by the European Space Agency. Their role is to provide high-resolution optical imagery to observe the Earth’s surface. As for the data we are working on, they are accessible from several sources but we have opted for access on Google Earth Engine ⁽³⁾ where the dataset from S2 is available.

However, it is not hard to imagine that turbidity does not appear directly on satellite images. To overcome this problem, researchers from BRGM (Geological and mining research bureau) in French Guiana ⁽⁴⁾ have built models that make it possible to establishing the link between reflectance and the effective turbidity of water. 

Above is the equation that corresponds to our expectations in terms of accuracy. However, XS1 and XS2 are the bands of the satellite that the researchers chose to create this model. Therefore we had to find bands of similar wavelength to reproduce these calculations for the Sentinel-2 satellites (in this case bands B3 and B4 respectively). 

And so we were able to calculate turbidity changes based on the reflectance from the S2 satellite images.

So what’s the bottom line?

We also used Google Earth Engine for our calculations. They offer a free web-based code editor that provides access to the data described above. Then, in Javascript, it is possible to use these data to display gradients of interest on a world map or to calculate particular values in a predefined area. 

The city that will be used as a guideline for the results is Venice. Italy is one of the countries most affected by the pandemic, but it is also a very touristy country. A symbol of mass tourism, an unavoidable stopover for cruises, Venice is, in spite of itself, a perfect example for this study.

To begin with, we chose to display a new spectral band on a map, created from the equation presented above. By taking images of autumn 2019 and early spring, we can indeed observe a significant difference between the hues of the sea around Venice before (fig.4) and after (fig.5).

This color difference is rather hard to perceive without being able to zoom in and out, so we considered the possibility of constructing a histogram of the evolution of turbidity. Before starting to build this histogram, we had to establish a precise zone around Venice, corresponding to the perimeter of the ships’ passage.

Thus we were able to make minimum, maximum and average calculations in this area, always taking dates before and during the pandemic. Here is a simple table before moving on to the histogram.

	Before the COVID-19	→ Decrease →	During the COVID-19
Min	491328	-22.3%	401583
Max	1782142	-29.6%	1374244
Mean	651981	-21.7%	535576

The histogram below provides a clearer visualization of the results of our study. This graph was designed by analyzing the value of the “turbidity” band for each pixel in the previously delimited area. By counting the pixels with values belonging to the same turbidity range, we can give a frequency at which this value returns, which makes it possible to plot a histogram.

Compared to the maps presented earlier in the article, it is much easier here to discern the marked decrease in turbidity. 

This variation in the curves coincides with the containment measures taken by governments, which have forced vessels (private or cruise ships) to stay in port.

It is this improvement in water quality coupled with a significant decrease in human activity that allows fish to come back in these waters and birds to wash in them. We can only hope that this will give us ideas on how to respect the aquatic environments we normally use on a daily basis.

The current crisis makes it possible to quantify the efforts needed to sustainably reduce the negative impact of human activity on the environment, and it is up to us to identify these measures in order to move forward intelligently in the future. 

Sources

1. : https://www.nature.com/articles/d41586-020-00885-w
2. : https://www.google.com/search=eauVeniseCoronavirus
3. : https://earthengine.google.com/
4. : http://infoterre.brgm.fr/rapports/RP-61632-FR.pdf