Automated Detection of Gridded Canal Networks in Veracruz, Mexico
Author(s): Kyle Urquhart; Wesley Stoner
Year: 2019
Summary
This is an abstract from the "SAA 2019: General Sessions" session, at the 84th annual meeting of the Society for American Archaeology.
The ancient peoples of Classic Period Veracruz employed a suite of strategies for agricultural intensification aimed at increasing agricultural yields and managing seasonal rainfall. One common strategy involved the construction of gridded canal networks with alternating raised field platforms which drained water in the wet season and retained it in the dry season using check-dams. Many of these gridded canal networks are clearly visible in high resolution satellite imagery, especially when using products of the near infrared bands. This paper outlines a method to automate detection and mapping of these features using Object Based Image Analysis (OBIA) through the software eCognition. A simple GIS model using slope, distance to streams or water sources, and vegetation indices is first constructed to narrow the search area. This is followed by a simple rule-based segmentation and classification process that aims to identify relatively regular and homogeneous patterns of alternating brighter and darker near infrared reflectance. The results show that the method serves as a statistically significant predictor of the location of gridded canals in Veracruz, and qualitatively the method serves as an effective mapping technique that may be of utility in other regions with similar features.
Cite this Record
Automated Detection of Gridded Canal Networks in Veracruz, Mexico. Kyle Urquhart, Wesley Stoner. Presented at The 84th Annual Meeting of the Society for American Archaeology, Albuquerque, NM. 2019 ( tDAR id: 450033)
This Resource is Part of the Following Collections
Keywords
General
Highland Mesoamerica: Classic
•
Object-based image analysis
•
Remote Sensing/Geophysics
•
Water Management and Irrigation
Geographic Keywords
Mesoamerica: Gulf Coast
Spatial Coverage
min long: -98.987; min lat: 17.77 ; max long: -86.858; max lat: 25.839 ;
Record Identifiers
Abstract Id(s): 23142