Why maps matter in our response to COVID-19

During the first half of 2020 when more than two-thirds of the world’s population was in lockdown, many of us were transfixed by a map. The alarming red and black display, produced by researchers at Johns Hopkins University, tracks the evolution of COVID-19, the most devastating virus of the past 100 years.

The map provides a real-time count of the number of people who are infected and killed by COVID-19 in almost 190 countries. Already viewed more than 200 million times, it is a reminder of how maps shape the way we see and engage with our world. As my new co-authored book, Terra Incognita, shows, remarkable improvements in geographic imaging and data sciences are helping us better prevent, respond and recover from crises, including disease outbreaks such as COVID-19.

Our understanding of and behavioural response to COVID-19 is profoundly influenced by maps. Since the disease first emerged in the Chinese city of Wuhan in late 2019, there has been a veritable outbreak of data visualizations charting its deadly spread. Some are more informative than others. When well designed and executed, maps can build awareness and shape smarter decision-making from the global to the hyper-local level.

Maps are, however, only as strong as the information that underpins them; their data sources and resolution will determine their usefulness. While there is no one way to spatially plot the diffusion and effects of COVID-19, a few maps stand out. Alongside the map produced by John Hopkins University, there are several prominent global dashboards documenting new cases and fatalities associated with COVID-19. One of the most authoritative is overseen by the World Health Organization (WHO), which provides a country-by-country update of cases. Working with the WHO, the global alliance for vaccines (GAVI) offers information on confirmed cases and deaths in 72 countries where it is active. A similar international dashboard offering even greater levels of granularity is HealthMap.

Companies with expertise in geospatial mapping such as Carto, ESRI and MapBox have also helpfully compiled expansive repositories of mapping tools. These firms have developed sophisticated software and accessible dashboards to help governments, businesses, health agencies and grassroots groups better respond and recover from COVID-19.

City-based COVID-19 mapping platforms are also proliferating. Local governments ranging from Ottawa and Toronto to Los Angeles and New York have developed systems to track infection, recovery and fatality rates, and local health capacities. Some companies are offering licensed products for urban authorities to develop real-time mapping systems using basic geo-imaging software, while others offer their wares for free.

There are even groups using these tools to help people identify infection risk by entering their zip codes. Citizen-led initiatives are also developing open source tools to help residents access up-to-date information. Among the lessons emerging from all of these efforts are the importance of ensuring a simple interface and prioritizing real-time data collection.

The underlying social and economic vulnerabilities of populations to infectious disease outbreaks are also being carefully documented. One example of this is RS21's urban health vulnerability index which measures the overall chronic disease burden of a US cities at the resolution of census tracts. It combines the health and demographic metrics to generate an aggregate score. RS21 draws inspiration from the Centers for Disease Control’s (CDC) social vulnerability index, which maps 15 variables such as poverty, lack of vehicle access and crowded housing.

Meanwhile, efforts are underway to capture the spread of COVID-19-related misinformation and disinformation online. A compelling example of this is the Infodemic Observatory produced by CoMuNe Lab. The platform reviews hundreds of millions of social media messages to interpret changes in public sentiment, the proliferation of bots and the reliability of information being shared online.

A related example is the Global Database of Events, Language and Tone (GDELT) which tracks, among other things, the most commonly mentioned news coverage of COVID-19. It tracks mentions of everything from government announcements of preparedness plans to how societies are internalizing, contextualizing and making sense of the pandemic. One of the first groups to detect the virus in late December 2019 – BlueDot – drew on data supplied by GDELT.

Several maps predict the potential spread of COVID-19 using a combination of computational methods. EpiRisk calculates the probability of exporting infected individuals by drawing on flight data. COVID Near You, a voluntary self-reporting platform designed in the US, helps individuals track symptoms. Likewise, Facebook is making anonymized mobility data available to show the extent to which physically distancing efforts are helping contain COVID-19, while SafeGraph tracks the percentage of the population staying at home (sheltering in place) every day.

A powerful platform tracking the response of governments was recently developed by the University of Oxford. Their government response tracker (OXCGRT) maps 17 metrics associated with containment and closure, economic and health system policies. The underlying data is combined into four common indices (0-100) to reflect the stringency of government action.

To help mapmakers improve the quality of their products, many private and non-profit entities are putting their datasets online for free. One particularly high-quality repository of data on health capacities is COVIDCareMap. The group provides detailed open-source information on hospital beds, ICU beds, bed occupancy and ICU bed occupancy, albeit only for the US. Another impressive inventory of country-level mapping efforts and their underlying data sources was recently made available by ESRI.

While not available free of charge, MapBox’s telemetry data provides more than 16 million location points a day on mobility patterns. Visualizing this kind of information can help policy-makers assess the extent of adherence to physical distancing and shelter-in-place measures.

The social and economic costs of COVID-19 are likewise being comprehensively documented using geographic imaging tools. A prominent example is the Urban Institute’s map tracking where low-income jobs are being lost at the metropolitan scale. International organizations such as UNESCO are tracking school disruptions – including closures affecting around 1.2 billion students – while the World Food Program is mapping school feeding programmes around the world.

Some research and consulting groups are bringing many of these methods together to help planners better respond to COVID-19. UrbanFootprint developed software to estimate how COVID-19 might affect at-risk populations and their relative access to basic services. They are also predicting which employment sectors are most likely to be affected, implications for public transportation and mobility, and where health authorities should deploy medical services and support.

Notwithstanding the explosion of real-time maps, there are still critical gaps in our knowledge. For one, the vast majority of digital tools are still geographically confined to North America, Europe and parts of Southeast Asia. While there are exceptions, there are still large swathes of the world – especially Africa, South Asia and Latin America and the Caribbean – that struggle to access even the most basic geospatial information on disease spread, health capabilities, social and economic risk factors, and their local resilience.

Equipping governments and service providers in lower and middle-income settings with reliable and easy-to-use mapping tools is critical to scale data-driven prevention, response and recovery. Ensuring that analytical platforms are sufficiently high resolution for city-level authorities is also essential. Dashboards tracking international and national statistics, while useful for high-level strategic planning, are not particularly helpful for a mayor, emergency response team or health provider on the frontline.

In order to help meet these needs, the SecDev Group is designing a data-driven platform to assist local governments and health providers, especially in lower and middle-income settings. The urban pandemic response and recovery index will be built in partnership with several cities in 2020 and distributed to inter-city networks such as the C40, Metropolis, the Global Parliament of Mayors, the Mayors Migration Council and the Global Resilient Cities Network in 2021. The platform will provide first responders and service providers with timely and actionable information on their health-related capacities, social and economic determinants of population health, and ways to foster local resilience.

When people explore maps, they may start to make connections that might have otherwise gone missing. The blending of geographic images with reliable data collected from public health agencies, research institutes, universities, companies and non-governmental organizations is changing the way many of us interact with the world. The visualization of real-world phenomena such as infectious disease can be at once dazzling, counterintuitive and enlightening. By removing uncertainties, they are also empowering, helping us move from a posture of alarm to one of action.