This whole problem of false or unknown negatives/absences of a spatial phenomenon. It reminds me of the Maxent modeling approach where you generate a bunch of pseudoabsences to compliment a limited set of presences. When I looked for an example with coronavirus and  Maxent, I found Here a different angle on the mapping of coronavirus problem using a bunch of machine learning approaches and climate data. THese folks think climate will constrain this thing.

On Thu, Apr 2, 2020 at 9:04 AM Tora Johnson <[log in to unmask]> wrote:
One more relevant item in today's New York Times: "For every known case of coronavirus, another five to 10 cases are out there undetected, a new study suggests." https://nyti.ms/2wNeNZz

~~~~~~~~~~~~~~~~~
Tora Johnson, PhD (she/ her)
Chair, Division of Environmental & Biological Sciences, Science Hall 107a
Director of GIS/ Associate Professor, GIS Lab & Service Center, Torrey Hall 223
University of Maine at Machias
[log in to unmask] ~ (207) 255-1214 ~ Twitter: @toradignity

Need to make an appointment with Tora during office hours? Go to http://bit.ly/toraofficehours

Like the UMM GIS Program on Facebook! https://www.facebook.com/ummgis
Like the UMM Science Division on Facebook! https://www.facebook.com/ummscience/
Looking for internships, jobs or grad school opportunities? https://sites.google.com/a/maine.edu/careers-umm-science/


On Wed, Apr 1, 2020 at 7:57 PM Tora Johnson <[log in to unmask]> wrote:
Dear Colleagues,

I hope you are all safe and well. I have been discussing covid-19 maps with fellow geographers around the state, and we believe we've identified an important problem. Nearly all of the maps we have seen show only the number of confirmed covid-19 cases. In spite of the fact that public health officials consistently say that these numbers represent only some of the cases and that the virus has likely already made its way into every county in Maine, we have heard many people, one-on-one and on social media, say that the virus is "not in my county" as an excuse for not practicing social distancing. Maps showing confirmed cases are probably part of the problem.

The folly of the "not-in-my-county" excuse was driven home for me over the last couple of weeks when my own 23 year old kid and their roommate came down with apparent cases of covid-19 in Portland. They were diagnosed by their doctors via telemedicine and never tested because they lacked other risk factors. Since this is currently the practice, clearly the number of actual cases is much, much larger than the number of confirmed cases, and the potential for spread across the state is much greater than confirmed case numbers would indicate. Both are recovering and are in quarantine at home, thank goodness.

From past studies on how people use and derive meaning from maps, we know that people often skip written information and derive a simplified message from a map or other such easily digested visualization. The maps of confirmed cases that are so common in the press and social media are likely no exception, so we believe many people are glancing at these maps and drawing erroneous conclusions. The geographers I have spoken with have agreed to begin including high-profile disclaimers in their maps to indicate that the number of actual cases is likely much larger and the virus is likely much more widespread than the number of confirmed cases would indicate. I'm writing to urge anyone creating such maps to do the same.

Currently, the state's official case count page includes the following disclaimer, however, it is in plain text and separate from the map: "This is likely an underrepresentation of the true number of cases in Maine since not all individuals are being tested. For individuals not considered to be at high risk, medical providers are currently advised to diagnose COVID-19 based on symptoms." So, it is likely that many people go to that page and glance at the map without reading the disclaimer. I have written to the CDC to recommend that they raise the profile of the disclaimer and place it on the map itself. 

I gather from what I have read that there is no way without more widespread and targeted testing to estimate the actual number of cases around the state (or anywhere else in the country). So, I understand that the number of confirmed cases is the only quantitative information available to report. Perhaps we can all work on improving the way this information is transmitted to the public with more careful map design.

I hope you find this information helpful. 

Be well!
Tora
~~~~~~~~~~~~~~~~~
Tora Johnson, PhD (she/ her)
Chair, Division of Environmental & Biological Sciences, Science Hall 107a
Director of GIS/ Associate Professor, GIS Lab & Service Center, Torrey Hall 223
University of Maine at Machias
[log in to unmask] ~ (207) 255-1214 ~ Twitter: @toradignity

Need to make an appointment with Tora during office hours? Go to http://bit.ly/toraofficehours

Like the UMM GIS Program on Facebook! https://www.facebook.com/ummgis
Like the UMM Science Division on Facebook! https://www.facebook.com/ummscience/
Looking for internships, jobs or grad school opportunities? https://sites.google.com/a/maine.edu/careers-umm-science/

--
You received this message because you are subscribed to the Google Groups "Maine Geospatial Institute-ALL HANDS" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [log in to unmask].
To view this discussion on the web visit https://groups.google.com/a/maine.edu/d/msgid/mgi-all-hands-group/CADa7MdsBC5GdDOKDajOj9m6ySpuuX%3Dk20jshKjUtcexmMk%2BVHg%40mail.gmail.com.


--
------------------
Peter Nelson, PhD
Assistant Professor
Biological Sciences and Environmental Studies
University of Maine-Fort Kent
Office: (207) 834-7683
https://peterrnelson.weebly.com/
----------------------------------------------------------------------- This list Maine GeoNews (GEOLIBRARY-L) is an unmoderated discussion list for all Maine GIS Users. If you no longer wish to receive e-mail from this list, you can remove your name and email address yourself at the this web address:

https://lists.maine.edu/cgi-bin/wa?SUBED1=GEOLIBRARY-L&A=1

Or, you may also request that your name be removed by sending email to:

[log in to unmask]