In their Commentary “Mapping disaster zones” (Nature 439, 787–788; 2006), Illah Nourbakhsh and colleagues rightly say that the global distribution of spatial data, especially high-resolution aerial photography, may raise issues of privacy violation. We at Louisiana State University used Google Earth extensively in the Emergency Operation Center to facilitate search-and-rescue missions immediately after Hurricane Katrina. There is no denying the utility this system provided. But, in the months after the hurricane, we used the markings left by these same search-and-rescue teams to highlight the danger of using geospatial technology.
A map displaying body-recovery locations in the New Orleans area was cut out of a local newspaper and turned into a Geographic Information System (GIS) layer, using census boundary information to guide the transfer as no streets and limited city landmarks were on the original. Once in the GIS, the central coordinates of each mortality ‘dot’ were extracted. A field team using the Global Positioning System checked the locations where bodies had been found against the search-and-rescue markings sprayed on houses. In many cases the dots, which on the map covered approximately one-and-a-half city blocks, revealed the location to be within an area of two or three neighbouring houses, with the exact house being identified in some cases.
What would have happened if a scientist had presented HIV cases, or nesting sites of an endangered species, on a similar city display — relying on an oversized point symbol to mask the real-world location? There is scant guidance for scientists to help preserve spatial confidentiality. Current standards often address only the required size for denominator populations in a thematic map display. Unless we in academia take the lead in expanding and enforcing a more rigid set of spatial display rules, especially for point data, we run the risk of an over-zealous tightening of data release and a protracted battle to again persuade those in power that a map can be used for the good of society.
About this article
Contributions to Correspondence may be submitted to firstname.lastname@example.org. They should be no longer than 500 words, and ideally shorter.
Geospatial cryptography: enabling researchers to access private, spatially referenced, human subjects data for cancer control and prevention
Journal of Geographical Systems (2017)