Using differential structure-from-motion photogrammetry to quantify erosion at the Engare Sero footprint site, Tanzania

• Published in: Quaternary science reviews 2018-10, Vol.198, p.226-241
• Main Authors: Zimmer, Brian, Liutkus-Pierce, Cynthia, Marshall, Scott T., Hatala, Kevin G., Metallo, Adam, Rossi, Vincent
• Format: Article
• Language: English
• Subjects: Change detection; Conservation; Differential structure from motion; East Africa; Engare Sero; Erosion; Footprint; Landscape evolution; M3C2; Photogrammetry; Present; Structure-from-Motion
• ISSN: 0277-3791; 1873-457X
• DOI: 10.1016/j.quascirev.2018.07.006

Hominin footprint sites are excellent sources of data that provide insights into early human physiology, anatomy, and social structures. They are also potential tourist attractions that are often situated in relatively under-developed parts of the world. Unfortunately, many footprint sites are also located in high energy environments and/or are pressed into poorly indurated sediments, which make them highly susceptible to erosion. This paper proposes a non-invasive and low-cost method employing Structure-from-Motion photogrammetry to quantify erosion rates in the absence of permanent ground control points. Using point cloud comparison algorithms between data collected at different times, it is possible to quantitatively analyze the locations, volumes, and rates of material loss. We applied this technique to several footprints within the Engare Sero footprint site in northern Tanzania to assess erosional change between 2010 and 2017. Our comparisons show that prints are vertically eroding at average rates ranging from 0.10 to 0.17 mm/yr with some localized areas experiencing much higher rates over shorter durations. We identify three primary modes of erosion: 1) flaking, 2) abrasion, and 3) boring of holes. Erosion appears to have occurred episodically with major events, such as flooding, separated by periods of relative stability. The methods presented here are valuable for paleoanthropologists to better understand how footprint erosion might adversely affect inferences regarding print-makers, and they are valuable for decision-makers, who can create conservation plans to better protect and maximize the utility of known hominin footprint sites. [Display omitted] •Erosional change can be detected using differential structure from motion.•Erosion presents in the form of abrasion, flaking, and boring.•Change detection is possible in the absence of permanent ground control.•Erosion may influence stature estimations of print-makers.