Trump’s COVID-19 ‘Risk Map’ will convey a false picture of safely for his base.

As part of President Trump’s ongoing campaign for a ‘packed churches’ Easter miracle, a dangerous trial balloon has been launched in the form of a ‘risk map’. Not known for his cartographic integrity — think Sharpie-Gate — Trump’s latest contrivance is county-by-county trojan horse ostensibly designed to inform the public as follows¹:

“Our expanded testing capabilities will quickly enable us to publish criteria, developed in close coordination with the nation’s public health officials and scientists, to help classify counties with respect to continued risks posed by the virus,” Mr. Trump said in a letter to the nation’s governors.

Like all Trump’s maps, this one will be plastered on foamcore — a fabricated amalgamation, part political dysfunction, part crisis mismanagement. The map’s ‘beautiful’ colors and ‘perfect’ gloss will coalesce as masterpiece of public policy subterfuge: instead of proactively informing vulnerable communities, it will simply aggregate counties along three classifications — ‘Low’, ‘Medium’ or ‘High’².

Cartographic aggregation is a long-standing tactic of propaganda maps, often deployed in crises to assuage a concerned public or hide unfortunate facts³. As we are reminded hourly, Covid-19 is decided not aggregated — it transmits person-to-person in close proximity. While other countries deployed early testing measures to literally map vectors of transmission before they became unmanageable outbreaks, the United States failed miserably in its early testing efforts. As a result, the only viable defense left is the lowest form of tech — literal distance, spacing and isolation.

What data is now available to the public is, indeed, aggregated to US counties. Even as they are useful to understand national spatial distributions, counties are inappropriate stand-ins for any meaningful epidemiological ‘hotspot’ mapping. Rather, counties excel when serving their original purpose as political geographies — coincidentally nestled neatly in blue states vs red states, urban blue vs rural red.

Both a NY Times data team⁴ and USAFacts⁵ are now publishing US Covid-19 county-level data on a daily basis. Mapping this data relative to urban vs rural geographies; US commute and economic megaregions⁶; and the distribution of hospitals in rural vs urban geographies — all point to critical vulnerabilities Trump’s ‘risk map’ will first paper over then exacerbate to dire consequences.

The Maps we see vs the Maps we should see — Part I

The typical Covid-19 map features proportional symbols — aka ‘bubbles’ representing relative size. Alternatively, a choropleth map — counties represented by sequential color denoting relative case counts. These two map types are effective at showing case distribution at a national scale:

Even as these thematic map types excel at case distribution at a national scale, each bubble or color only represents the aggregate count at the ‘center’ of the county. The actual location of ‘hotspots’ within counties remains unknown. Further, these maps are single variable thematic maps; they capture one dimension of Covid-19 — the hazard — not vulnerabilities and certainly not cumulative risk.

The Maps we see vs the Maps we should see — Part II

As we’ve watched New York City’s case rate soar, the instinctive lesson of Covid-19 appears to be this is an urban problem. On its face, true indeed. At left, US Census urban areas; at right, proportional representation of confirmed cases as of March 26, 2020:

Overlay of US Census urban areas + proportional representation of total cases.

However, like counties, urban areas are composed of structured, nestled geographies — US Census tracts. Even as humans live within both urban and rural geographies, importantly they move across these geographies on a daily basis. This movement — commutes — determines ‘megaregions’ of economic activity. Severely curtailed by Covid-19, its this movement that Trump seeks to ‘release’ in order to deliver the ‘ Easter Miracle ‘. At left, the ‘megaregion’ commute trajectories; at right, again, total confirmed cases per county:

Overlay of economic ‘megaregions’ + Proportional representation of total cases.

Even as these maps denote the essential distribution of Covid-19 relative to human activity defined as economic ‘megaregions’, cumulative risk is unknown as vulnerabilities have not been adequately mapped. Rarely seen are the Covid-19 maps that consider vulnerable populations — demographic cohorts most susceptible to the virus. These are typically older individuals and communities with prevalent underlining medical complications. If these vulnerable populations become infected in significant number, the last resort will be hospitals with ventilator capacity. Unfortunately, medical facilities across the United States — most acutely in rural geographies — have been shuttered en masse, resulting in a compounding infrastructure vulnerability atop underlining social vulnerabilities.

Robust geographic distribution of existing hospital locations can serve as a proxy of exposure resiliency. Conversely, sparse hospital capacity operates as an infrastructure vulnerability thereby increasing overall risk. Mapped below are existing hospital locations, each represented as a black dot. At the county geographies, the count of hospitals is normalized to a rate of hospitals per 100k populations. Finally, the rate of confirmed cases per 100k populations is mapped to underlining hospital coverage:

Overlay of hospitals, rate of coverage + rate of confirmed cases per 100K populations.

The results of this mapping demonstrate complicated exposures across both urban and rural geographies. While Covid-19 rages in urban areas, the rate of hospital coverage actually trends lower than many rural counties. On the other hand, many rural counties have literally one hospital, most consistently in the midwest. Following the immediate crisis in urban geographies — especially across New York City — underlining cumulative risk is clearly prevalent in many rural geographies. As economic megaregions have only grown over the past decade, fusing urban and rural geographies through economic symbiosis, rural areas may face acute and mounting risks exacerbated by location, social and infrastructure vulnerabilities.

The complex interplay of hazard, resiliency, exposure and final risk across local geographies simply cannot be captured by a highly aggregated ‘risk map’ of divvied up counties. Covid-19 is not aware nor obedient of containment by political geography. While Trump will undoubtedly tout ‘safe zones’ outside Covid-19’s early urban footprint, the virus will create its own map, feeding on vulnerabilities made worse by political division and crisis mismanagement.

Stephen Metts is a GIS analyst and instructor based in New York State. His research interests covered in this article include public policy, public health and risk mapping.

GIS Analyst & Instructor | Shale Gas Impacts, Environmental Justice & Climate Change Issues