Interpreting and Georeferencing the Concept of “Near” in Biodiversity Records

Georeferencing historical biodiversity specimens is a difficult but necessary task to bring data accumulated in the course of past scientific efforts into full relevancy for modern use. Textual locality descriptions vary widely, and are prone to error involved in interpreting brief summaries of place names, interpretation of incomplete or unclear terms, and even simply understanding handwriting. To address this challenge, each type of locality description requires particular georeferencing methods to maximize precision and accuracy of resulting coordinates and uncertainty. Some of the current "best practice" methods concerning textual descriptions of proximity to a locality (“near” a locality) are arbitrary, restrictive, or undefined. In this thesis, I explore these challenges, and develop and provide new methods for georeferencing such challenging locality descriptions, using Voronoi polygons to define rigorously and systematically the idea of “near.” Voronoi polygons literally define the geographic space nearer to a given point than any other point in a collection, making them ideal for methodically determining the shape of these amorphous locality descriptions. I explore the application of these approaches to a variety of georeferencing challenges, and discuss further improvements that may also become possible.

Grantor dc:publisher

University of Kansas

Year dc:date.issued

2023

Author dc:creator

• Campbell, Peter D.

Advisor dc:contributor.advisor

• Peterson, Andrew T

• Ecology
• Geographic information science and geodesy
• best practices
• biodiversity collections
• coordinate uncertainty
• georeferencing
• open-source
• Voronoi polygon

Statement dc:rights

• Copyright held by the author.

Language dc:language.iso

en

Dc Identifier Other

http://dissertations.umi.com/ku:19091

OAI identifier oai:identifier

oai:kuscholarworks.ku.edu:1808/38801