The Benefits of Short-Wave Infrared Imagery for Archaeological Landscape Analysis: A Case Study from Easter Island, Chile

Author(s): Dylan Davis; Carl Lipo

Year: 2017

Summary

The use of multispectral imagery is particularly effective for studying the archaeological record of Rapa Nui (Easter Island, Chile) due to the lack of vegetation and the fact that record is composed of surface distributions of rock features. Flaws (2010) has demonstrated that WorldView-2 multispectral imagery that includes the NIR band can be used to identify "lithic mulch gardens," a key component of prehistoric Rapa Nui subsistence strategies. Recently, the availability of WorldView-3 satellite imagery data offers researchers access to short-wave infrared (SWIR) bands, which provide significant additional information about moisture content and mineral composition. Unfortunately, this imagery is currently only available at a much lower spatial resolution than NIR images. Here, we evaluate whether or not this new source of SWIR imagery can be used for measuring "lithic mulch garden" features despite its significant resolution difference. Comparing the results of Flaws (2010) analysis of with the results obtained using maximum likelihood classification analysis of SWIR WorldView-3 imagery we found markedly similar classification accuracy, despite having a significantly lower spatial resolution. This result suggest that SWIR may provide a new tool for researchers interested in questions of prehistoric land-use that will become increasingly powerful as greater spatial resolutions become available.

Cite this Record

The Benefits of Short-Wave Infrared Imagery for Archaeological Landscape Analysis: A Case Study from Easter Island, Chile. Dylan Davis, Carl Lipo. Presented at The 81st Annual Meeting of the Society for American Archaeology, Vancouver, British Columbia. 2017 ( tDAR id: 428924)

This Resource is Part of the Following Collections

Keywords

Geographic Keywords
Oceania

Spatial Coverage

min long: 111.973; min lat: -52.052 ; max long: -87.715; max lat: 53.331 ;

Record Identifiers

Abstract Id(s): 14512