The world’s leading publication for data science, AI, and ML professionals.

Why Spatial is Special in this age of Coronavirus

No. I don't mean Dashboards with Maps but more profound ways.

Photo by Patrick Assalé on Unsplash
Photo by Patrick Assalé on Unsplash

I am sure you noticed the prevalence use of maps in almost every COVID-19 dashboard or data visualization. And if you have already come across the famous John snow’s 1854 – The father of epidemiology – cholera map of London, it should not come as a surprise how spatial data analysis can aid in fighting epidemics.

However, geography or spatial data science is helping the fight against this pandemic in more subtle ways that are not at the spotlight, like the ever-present dashboards and other data visualization of COVID-19 cases.

I mean, we tend to forget the mobility patterns that are readily available across the globe that can enhance studying and predicting the spread of the virus. We also have more frequent satellite revisits around the world to monitor human activities. Let us some innovate uses and applications of spatial data science in response to COVID-19 Pandemics.

Mobility under Quarantine

Despite the privacy concerns over Smartphone tracking data, their usage under COVID-19 gives us unparalleled insights into people’s movement under the social distancing measures. In fact, countries utilized ¹ these technologies and mobility dataset available to study or predict the spread and size of epidemics. And more recently, the European commission provisioned to employ mobility data from phones to tackle COVID-19 pandemics.

Aggregated and de-anonymized datasets have been made available through different initiatives, for example, Facebook’s data for good and Google’s COVID-19 Community Mobility Reports.

Mobility is an essential piece of information we need in order to respond to pandemics. As studies² show using these kinds of datasets like online social networks prove to be useful to epidemiologists and others hoping to predict the spread of diseases such as COVID-19 as a study which uses Facebook data suggests.

Social Connectedness Index – Facebook

Take, for instance, Facebook’s Social Connectedness Index, where we can understand general movements patterns of people and also model the probability and predict the chance that a disease will be spread by human-to-human contact.

The Social Connectedness Index measures the strength of connectedness between two geographic areas as represented by Facebook friendship ties. These connections can reveal important insights about economic opportunities, social mobility, trade and more. (Facebook Data For Good)

Social connectedness example from Italy.
Social connectedness example from Italy.

Community mobility reports – Google

Google has also made Community mobility reports available online, ready to be used during the COVID-19 period for most countries in the world. The measured the decrease or increase of 6 types of mobility (Retail & recreation, Grocery & Pharmacy, Parks, Transit Stations, workplaces, and Residential. As you can see from an extract of Google’s community mobility report from London, all these groups show a decrease of mobility except residential areas where we can see an increase in time spent at home.

Thus, mobility data can not only enable us to understand human mobility and behaviour under these hard times but also helps model and study the spread of COVID-19 and how social distancing prevention measures affect the spread of the disease. These datasets prove to be essential to tackle and reduce the spread of this pandemic.

Birds’ Eye View from Satellite Images

Seeing the Coronavirus impact from space is also another spatial data science front that illustrates the profound effects this epidemic has on the world economy and societies. The satellite imageries can reveal from bird’s eye view on how life stops under this pandemic. As many gatherings are cancelled, empty places can easily be seen through satellite imagery.

Great Mosque of Mecca, or Haram Mosque, in Mecca, Saudi Arabia. Left (February 14), Right (March 3) - Source
Great Mosque of Mecca, or Haram Mosque, in Mecca, Saudi Arabia. Left (February 14), Right (March 3) – Source

The study of where we see the highest drop-in activities and applications of deep learning and AI to detect and classify the socio-economic impact of the coronavirus on these places are vital to the long recovery process that awaits us.

Climate and Covid-19

Another significant subtle insight from satellite imagery is how the complete stop of human movement affects the environment. It has been a great time to study climate dynamics for climate researchers. Thanks to the launch of satellite missions that monitor the atmosphere, we can now glean and explore the human footprint and the climate. Take, for example, the publicly available sentinel 5 data from TROPOspheric Monitoring Instrument (TROPOMI) instrument, co-funded by the European Space Agency (ESA) and The Netherlands.

Nitrogen dioxide emissions in Spain during COVID-19 and last year 2019 - Source
Nitrogen dioxide emissions in Spain during COVID-19 and last year 2019 – Source

Conclusion

Spatial analysis done so far are only the tip of the iceberg, and there is a vast potential to use both mobility dataset as well as Earth observation data to reveal hidden patterns and insights on socio-economic changes under this period. Open and Free spatial datasets available around the world have much more promising potential for studying and mapping out changes in different modes transportation, connectivity and mobility patterns, as well as gaining insight into the real world dynamics of social distancing and quarantine.

The subtle and less subtle of deploying spatial component into the fight against COVID-19 can elicit so many unanswered questions during the coronavirus pandemic. Why is the mortality rate so different in different parts of countries around the word? How do geography, weather, population age distribution, lifestyle and other conditions affect the spread or containment of the coronavirus?

Note from the editors: Towards Data Science is a Medium publication primarily based on the study of Data Science and machine learning. We are not health professionals or epidemiologists, and the opinions of this article should not be interpreted as professional advice. To learn more about the coronavirus pandemic, you can click here.

Reference

[1] http://pages.stern.nyu.edu/~jstroebe/PDF/SCI_and_COVID.pdf

[2] https://science.sciencemag.org/content/early/2020/03/25/science.abb4218


Related Articles