Authors: Daniel Griffin*, University of Minnesota, Colin McFadden, University of Minnesota, Kate Carlson, University of Minnesota, Daniel Crawford, University of Minnesota, Grace Hanson, University of Minnesota, Malik Nusseihbeh, University of Minnesota, Samantha Porter, University of Minnesota, Jessica Thorne, University of Minnesota, Matthew Trumper, University of Minnesota
Topics: Biogeography, Quantitative Methods
Keywords: tree rings, geochronology
Session Type: Virtual Paper
Start / End Time: 11:10 AM / 12:25 PM
Room: Virtual 30
Presentation File: No File Uploaded
Ultra high resolution imaging is becoming standard across the sciences and should be a priority for dendrochronology. Large format scanners including the Epson XL are insufficient for resolving micro rings and the myriad anatomical structures of increasing scientific interest. Current software is unsatisfactory in terms of cost, user experience, data management flexibility, and capacity for handling large file sizes. We argue for a new paradigm and present a framework that unifies gigapixel resolution panoramic macro photography of physical specimens, a metadata compliant cloud hosted digital asset management system, and an internet accessible toolbox for dating, annotation, and increment measurement. During development, these platforms have proven advantageous compared to other image analysis approaches and the analog stage micrometer. Accurate and precise measurement data development is more efficient in terms of time and quality control procedures, particularly with new trainees and the subjective demarcation of sub-annual ring boundaries. This system is well suited to distributed and remote research collaboration, teaching, and outreach. Toward the end of quantitative wood anatomy, preliminary results with machine learning and computer vision are encouraging. We are developing options for affordable image acquisition hardware, and our software packages are intended to be open source, free, and community driven. Our vision is to leverage emerging technologies to produce an internet accessible digital archive to complement entire collections of physical tree-ring specimens. In light of recent debates about missing rings and data authenticity, we argue that our approach represents a critical and overdue step towards open science standards in dendrochronology.