The world has not seen a global health crisis like the COVID-19 pandemic since the Spanish flu in 1918.
Unlike 1918, however, the overwhelming amount of information available to the general public today can be daunting. And the speed with which it is disseminated online carries a distinct set of challenges around squaring what people are observing locally firsthand with what they’re seeing reported through various media. To make matters worse, since most data are reported through an interpretive lens, inferences are often contradictory or even misleading. This has made an already precarious situation all the more difficult to assess and plan around as regions begin loosening lockdown orders.
Putting COVID-19 Data into a Geospatial and Integrative Framework
When confronted with volumes and varying sources of data, the ability to aggregate and analyze information as it changes becomes increasingly important. Most media emphasis to date has been around tracking the number of daily reported COVID-19 cases and deaths to see if social distancing measures are having the intended effect of “flattening the curve” associated with infection rates.Figure 1 is an example of the type of graph commonly shared to track the daily increase in COVID-19 cases reported nationwide. After a steady rise in the number of cases recorded through March, the increase slows by mid-April and now we may be starting to see an average decline in new cases.
By leveraging publicly available data through state and federal authorities, we assembled a series of maps below that tell a story of a virus that has spread at alarming rates in some parts of the US, while moving much slower into others. Plotting available data in a geospatial context offers a glimpse of the areas hit the hardest so far. The added ability to layer various datasets in context together on a map provide further potential for gaining insights and spotting trends that may otherwise get overlooked.Figure 2 provides an example of how to view a geospatial distribution of COVID-19 deaths against a layered background representing population density by US county.
A Contemporary Approach to Visualizing Trends
Similarly, we can contextualize the results of COVID-19 testing by geo-locating confirmed positives on a per capita basis. For example, the following map depicts confirmed COVID-19 cases per 100,000 people, reflecting a relative infection rate when normalized for county population (Figure 3).This visualization suggests a more nuanced distribution of the geographic impacts associated with the virus. By normalizing the rate of infection in relation to population, we get a different glimpse of the challenges the pandemic poses to a broader cross-section of the country. In darker areas the impact has been quite pronounced. But those living in lighter shaded areas may not know anyone who’s had the disease. The annotations reflect local testimonials from different regions of the country.
Adding Insights to Your Daily Work
It is hard to think of an industry that has not been impacted by COVID-19. Rendering these maps as base layers can place the impacts in context with operations. Normalized county population has particular interest to the energy sector. With operations concentrated in rural and less densely populated areas, visualizing data against this backdrop is useful for evaluating on-site conditions and planning future activities.
To illustrate, Figure 4 highlights counties with oil and gas drilling activity. Counties with a light gray outline showed activity on April 4th, and those with a black outline on April 27th. The small green rig symbols show where rigs were located as of April 27th.Some rural counties with low population have yet to observe COVID-19. That is likely to change within a few weeks. We can gain further insight on the timing of impact in different regions by comparing against a map reflecting when COVID-19 cases were first reported.
Figures 5 and 6 combine to show how COVID-19 has spread like a wave over the country. Pink shaded areas represent the earliest cases and light blue represents counties reporting first cases later. Early on, only a few counties detected the virus. But by the time 62 days had passed, 44% of counties reported cases. And by 103 days, confirmed cases spread to all but 10% of US counties.
COVID-19 Information Continues to Evolve
Based on these geographic renderings of the available COVID-19 data, it is understandable why many are unsure how to interpret the pandemic’s impacts around the country. To date, 90% of COVID-19 cases have been reported in counties representing half the country’s population. It is even more pronounced with the death statistics: 90% of deaths are in counties with 35% of the population. That’s a striking difference in experience depending on where one lives. Population density isn’t the only variable at play, however. The maps showing infection rates adjusted for population indicate a more nuanced distribution of the virus. More densely populated areas may be experiencing the worst now. But seeing how cases have spread geographically over time, every region will be challenged in the coming months.
Like what you see here? Our proprietary Change Detection feature will also allow you to detect any variance in the datasets when they are updated. Please note we are able to integrate almost any type of data in our platform. If you have other datasets you think might be helpful to see in conjunction with what we’ve aggregated here, please contact us at: support@PetroDE.com.