CSCI-E-79 The Art & Design of Information

Learning from the Curve

COVID-19 has put the world on pause. As cities across Europe implement lockdown measures, the impact on pollution is a breath of fresh air.

The COVID-19 Story

The WHO began surveilling Europe for COVID-19 on January 27th, 2020. As the number of cases ticked upward, cities and countries across the continent implemented measures for containment — at first, reluctantly, then swiftly and out of necessity. The magnitude of the resultant social and economic impact is difficult to overstate and will ripple into the future in somewhat predictable but largely unknowable ways. Data on how air quality has changed, however, tells a clear story of positive difference.

It follows that when travel, even on local scales, is greatly reduced, there would be a decrease in emissions. But what does this look like — how big is the change and how immediate is the effect?

Cities in Focus

A visual analysis of measurements of NO2 taken in six cities across Europe shows how concentrations of this gas have changed post-lockdown. NO2, a product of the combustion of fuel, is an emission for which the two most common sources are vehicles and power plants. The overall impact for each locale varies depending on factors which range from the more abstract (e.g. community ethos and behavioral tendencies) to numerical observation (e.g. concentration of emissions preceding lockdown).

0 μg/m3
N02 Concentration
200 μg/m3

The Monthly Difference

In general, Europe has made concerted efforts to reduce its environmental footprint and succeeded. Supporting this, plotting daily levels of NO2 from January 1st through April 30th shows that there is an observable reduction from 2018 to 2019, and from 2019 to 2020. There is a wavelike up-down periodicity which appears to occur year after year, perhaps indicating traffic fluctuations that correlate to days of the week. Many such relational and pattern-based observations surely exist, but the focus of this analysis is what occurs at and as a result of lockdown.

On March 9th, Milan became the first major European city to implement lockdown measures. Copenhagen followed suit on March 13th, with Madrid close on its heels on March 14th. Paris went into lockdown on March 17th, and then London on the 23rd. Stockholm did not instate any official lockdown.

It is clearest in the case of Madrid, and nearly as noticeable for Milan and Paris, that after issuance of recommendations for extreme social distancing, the level of NO2 sees an immediate and marked drop that is sustained. Copenhagen and London show a slight, perhaps contestable reduction in concentrations of NO2, while Stockholm, the only city of this set which did not implement lockdown, exhibits no real difference in its NO2 trends for January to the end of April.

The visualizations below show daily range and average NO2 concentrations from January through April, 2018-2020. Use the filters to highlight specific cities, years, and metrics.







The Hourly Difference

This post-lockdown reduction in concentration of NO2 is further supported by comparing hourly data from one day that falls before and one that falls after lockdown. Starting with two Mondays (to eliminate the suspected variance in traffic trends between days of the week), a comparison of the hourly averages over the course of a day shows that at almost every hour, there is a reduction in concentrations of NO2. As would be expected, this bolsters the conclusion that there is a reduction in emissions. This temporal scale also makes it possible to observe the emissions peaks in the morning and evening, and these visible trends which map to rush hours lend additional insight, primarily into how responsive NO2 concentrations are to the number of cars on the road at any given time: how quickly levels rise, how long NO2 stays in the air, and how quickly levels fall.

All cities in focus (save Stockholm) went into lockdown mid-March, so March 2nd and March 30th highlight a consistent pre- vs. post-lockdown view. Explore your own custom timeframes using the date selectors below.







The Yearly Difference

Visualizing the same data with an eye trained on net impact continues to support what we already know to be true and produces, for those more interested in succinct statement, an overall percentage decrease. The change for London is 15%, which is not an insignificant year-over-year reduction but appears modest compared to Copenhagen’s 47% and Paris’s 57%, and is a mere one-fifth of Madrid’s 65%.

Hover over the visualization below to reveal the daily year-over-year difference in average NO2 concentration.

Stockholm no lockdown London 15% Paris 57% Madrid 65% Copenhagen 47% Milan 31% January 4th-65.6 January 7th-78.8 January 10th54.2 January 11th-20.6 January 12th85.4 January 13th-20.9 January 14th16.3 January 16th-59.8 January 20th93.2 January 22nd-25.0 January 24th18.3 January 26th-40.8 January 27th7.0 January 29th-37.8 February 14th-48.6 February 17th-37.6 February 19th15.8 February 24th-27.9 February 27th-74.0 March 1st43.0 March 2nd-34.6 March 3rd1.9 March 6th-32.8 March 7th18.1 March 11th15.8 March 14th-28.8 March 18th37.8 March 23rd-77.3 March 27th-33.3 March 29th-45.3 March 30th23.3 April 1st-61.0 April 5th-26.4 April 9th6.1 April 12th-42.3 April 18th-50.0 April 22nd8.2 April 25th-28.0 April 28th2.0 April 30th-14.6 January 2nd5.9 January 3rd-29.3 January 9th29.2 January 11th-31.9 January 13th21.3 January 15th-10.9 January 16th3.2 January 18th-30.8 January 21st-29.7 January 25th-18.3 January 30th-15.6 February 1st-30.0 February 3rd12.2 February 4th-27.8 February 5th1.3 February 8th-14.7 February 12th-22.0 February 14th9.7 February 16th-38.1 February 18th-24.4 February 22nd-31.5 February 25th-18.3 March 1st-53.1 March 3rd24.0 March 8th-5.8 March 9th24.5 March 11th-33.4 March 15th-20.9 March 16th14.5 March 19th-28.1 March 25th-12.8 March 29th-28.1 March 31st-0.8 April 1st-35.7 April 3rd-35.3 April 7th-32.3 April 8th2.0 April 10th-25.2 April 10th-25.2 April 15th-35.2 April 17th-26.8 April 20th-46.3 April 23rd16.1 April 26th-53.2 April 30th-26.5 January 3rd76.5 January 4th-47.7 January 6th-42.2 January 8th13.4 January 8th13.4 January 14th-75.9 January 19th-39.0 January 21st-29.3 January 23rd12.3 January 24th-17.4 January 25th37.0 January 27th47.1 January 29th-14.9 January 31st1.1 February 1st-13.8 February 3rd3.7 February 5th-54.9 February 7th-41.9 February 10th15.2 February 13th-42.6 February 15th-45.3 February 18th-23.2 February 23rd-37.6 February 24th7.0 February 26th-52.7 March 1st-51.5 March 5th-14.1 March 9th-14.1 March 11th16.2 March 15th-79.8 March 22nd-53.5 March 25th-34.0 March 28th-52.5 April 2nd-41.8 April 5th-45.3 April 9th-39.9 April 12th-56.5 April 16th-38.7 April 19th-23.8 April 23rd-32.0 April 27th-43.2 April 30th-41.7 January 2nd5.1 January 3rd-12.8 January 5th9.8 January 10th-19.4 January 12th-3.9 January 13th16.0 January 15th-21.9 January 18th-37.3 January 20th1.3 January 21st-28.0 January 22nd2.0 January 24th-16.9 January 29th-25.2 February 1st -55.0 February 4th-37.1 February 6th3.9 February 13th-45.8 February 16th-77.5 February 22nd-78.7 February 27th-67.1 March 4th9.8 March 6th-23.7 March 8th-19.0 March 10th5.4 March 12th-20.1 March 13th3.7 March 17th27.4 March 20th-63.9 March 22nd-65.9 March 30th-76.5 April 4th-38.2 April 6th-36.5 April 8th-16.3 April 10th-17.7 April 13th-39.0 April 19th-74.6 April 22nd-48.5 April 26th-28.0 April 27th1.4 April 28th-11.0 April 29th13.6 April 30th-44.6 January 2nd-18 January 6th-34.6 January 9th-32.7 January 11th-42.2 January 14th-29.3 January 17th-17.8 January 21st28.7 January 25th-27.8 January 28th14.2 January 31st-32.7 February 4th33.7 February 6th38.0 February 10th6.8 February 13th22.3 February 15th-34.6 February 19th22.6 February 22nd-20.1 February 25th-28.9 February 27th-16.6 March 1st-22.4 March 3rd47.5 March 6th36.2 March 8th-10.7 March 9th13.8 March 13th20.0 March 14th-7.0 March 16th41.2 March 18th-27.3 March 20th-7.5 March 23rd12.6 March 24th-6.4 March 25th21.8 March 26th-11.9 March 30th-33.4 April 2nd20.3 April 5th-15.1 April 6th18.0 April 10th25.8 April 13th-12.9 April 17th-24.8 April 22nd-27.6 April 25th8.4 April 27th6.6 April 30th-29.3 January 3rd-25.6 January 8th-25.6 January 9th6.7 January 12th-20.9 January 15th-31.0 January 18th-27.9 January 21st-36.3 January 24th-35.8 January 27th5.0 January 28th-5.6 January 31st-10.2 February 3rd30.6 February 5th-6.6 February 6th8.9 February 8th-5.1 February 10th0.7 February 12th-14.8 February 14th32.2 February 17th-16.2 February 19th4.2 February 22nd-30.3 February 24th22.1 February 26th-24.7 February 28th12.4 March 1st-22.8 March 2nd9.2 March 5th-3.5 March 7th6.8 March 8th-5.9 March 9th2.7 March 12th-10.5 March 13th8.8 March 15th-16.8 March 16th2.0 March 19th-16.2 March 22nd-13.0 March 24th8.2 March 28th-15.0 March 30th8.7 April 5th-7.4 April 7th5.4 April 12th-21.6 April 19th-27.7 April 22nd10.4 April 26th-12.5 April 28th4.4 April 30th-20.3 decrease since lockdown X% difference in average concentration over previous year (μg/m³) decrease increase

Lessons of Lockdown

It is obvious that this level of reduction in human activity is not reasonable or sustainable. At some point, with many of us hoping soon, things must return to normal. But isn’t there a degree to which we can carry these emissions reductions forward? What can we choose to change, on a manageable scale, about the way we live?

The concept of being better stewards of the environment is not a new one. We are well-versed in recommendations for improvement: conserve water, reduce waste, recycle, recycle, recycle. But it is easy to lose sight of how these small, voluntary changes are wise investments in a shared future.

Contending with COVID-19 has pushed the world into a pause which, under normal circumstances, would have been optimistic as a proposition, and impossible in execution. But we are here with the numbers in front of us, and the possible takeaways are many. In order to reduce the number of cars on the road at any given time, a well-maintained and reliable public transit system within a city is a must. Furthermore, the connection of cities in a national public transit network is something we can work towards and hope for. In restructuring the supply chain, we should aim to increasingly source food, commodities, and raw materials locally, thereby minimizing the need for long-distance shipping.

We would do well to educate ourselves about the history of our immediate communities, to become versed in our local politics. Write to, call, petition policymakers — make it a priority to vote. Infrastructural aims guided by shifts in policy are important, but the time between personal action and visible change might unfold over months, years, or decades. In light of the time it takes for these changes to take effect, persistence is key. These ends may start to feel distant, but each individual can take heart: there are small, no less impactful decisions for the day at hand, the effects of which are immediate and lasting.

Once a day, once a week, or even as sparingly as once a month, ask yourself: “Do I need to drive myself to get where I am going, or can I travel some other way?”

Enjoy the fresh air responsibly...

Our data-driven fashion face masks turn air quality visualizations into wearable stories, with printed patterns that reveal the real-world effects of COVID-19 lockdowns. Our goal is to provide a beautiful product that encourages you to keep yourself and those around you safe.

Please note: masks are non-medical grade and do not include N95 filters. These are for personal use only.

data driven face mask
data driven face mask
Buy these masks

Data analysis, design and development by:
Douglas Sanchez, Rebecca Lantner, and Grace Park

Special thanks to Zona Kostic for leading an inspiring course,
Ning Chen for her input as TA, and last but definitely not least,
Jared Jessup for committing many hours to helping us debug
and encouraging us to work hard and think big.