Three-Dimensional Recording of Archaeological Remains in the Altai Mountains

Article excerpt

[ILLUSTRATION OMITTED]

Introduction

The Altai Mountain region (south-west Siberia) is amongst the archaeologically richest of Central Asia (Figure 1). Numerous burial mounds, ritual surface monuments and rock art sites, dating from the late Neolithic (3200 BC) onwards are scattered across its valleys, highlands and steppes. Insights into their density and distribution are limited despite a century of archaeological excavations, but recently, well-directed surveys have shown the full archaeological potential of this broad region (Allard & Erdenebaatar 2005; Bourgeois et al. 2007; Krupochkin 2009; Wright et al. 2009; Jacobson-Tepfer & Meacham 2010). Unfortunately, infrastructural growth, intensive agriculture and increasing tourism are putting pressure on the archaeological heritage (Plets et al. 2011a). Moreover, the planned pipeline through the Altai illustrates the necessity of thorough mapping and documentation of archaeological sites (Plets et al. 2011b).

[FIGURE 1 OMITTED]

Since 1999, extensive areas of the Russian Altai have been surveyed by Ghent University, using satellite imagery and Global Navigation Satellite Systems (GNSS) (Gheyle et al. 2004; Bourgeois et al. 2007). Precise GNSS measurements, photographic documentation and detailed descriptions have been collected for more than 15 000 recorded monuments. This procedure has proved adequate for most monuments since they belong to an established typology. However, exceptional monuments with striking variations, decorations, or a complex typology were also encountered. These monuments deserve to be recorded and mapped at the highest level of detail, to enable the acquisition of all possible relevant information needed for current and future research. Here the default descriptions, paper sketches and standard digital photographs have proved to be insufficient.

Technical advances in the field of three-dimensional (3D) recording using photogrammetry or terrestrial laser scanning have begun to fill this gap (Lambers et al. 2007; Lerones et al. 2010; Hendrickx et al. 2011). Many of these are time-consuming and their complexity or cost hampers a fast and straightforward implementation during field research. Moreover, a high level of expertise and expensive hard- and software is required. By contrast, recent developments in the area of computer vision show great potential for fast, flexible and detailed documentation of heritage, without specialised and expensive instruments (Simpson et al. 2004; Sanz et al. 2010; Doneus et al. 2011; Verhoeven 2011). The aim of this paper is to assess the potential of a recently developed computer visionbased package, PhotsoScan (AgiSoft LLC), to document the more complex archaeological monument types of the Russian Altai (i.e. rock art and stone surface monuments) in a precise and cost-effective manner, without compromising flexibility or efficiency.

Archaeology of the Altai Mountains

For millennia, the Altai Mountains have been an important transitional zone between the Mongolian and Kazakh steppes. Consequently, the region has an extraordinary number of archaeological monuments and rock art sites, assignable to many cultures. Moreover, some of the burial sites, mainly dating from the Scythian period (800-200 BC), are located in permafrost, resulting in well-preserved wooden objects, textiles, leather ornaments and even the remains of sacrificed animals and mummified humans (Rudenko 1960, 1970; Molodin 1992; Polosmak 1995).

The archaeological heritage not only has an important scientific relevance, but also supernatural significance to the indigenous Altaians. The 2003 earthquake, for example, was blamed locally on the excavation of the Scythian burial mound of the so-called ice princess of Ukok (Halemba 2008), and led to an ongoing dispute between archaeologists and the indigenous population that has impeded excavations ever since. Thus, the conservation of the archaeological heritage of the Altai can also be considered socio-culturally imperative. …