Spectral Discrimination of Archaeological Sites Previously Occupied by Farming Communities Using In Situ Hyperspectral Data

Abstract

<jats:p>This study investigates the ability of field spectra measurements to discriminate between soils from non-sites (natural soils) and from archaeological sites, such as middens (rubbish-dumping areas) and animal byres. First, we tested whether there is a difference in the concentration of elements between different soil types using analysis of variance while random forest (RF) and forward variable selection (FVS) methods were used to select important soil elements for the classification of the archaeological sites. In the second approach, we evaluated the ability of field spectroscopy reflectance measurements to discriminate among nonsites, middens, vitrified dung, and nonvitrified dung byres. The guided regularised random forest (GRRF) was used to identify important wavelengths for the discrimination of abovementioned archaeological and nonarchaeological soils. Thereafter, the selected soil elements and wavelengths were used as input variables in the RF classification algorithm to classify the nonsites, middens, vitrified dung, and nonvitrified dung. The findings reveal that there is a significant difference in the composition of chemical elements and spectral signatures of nonsites, middens, vitrified dung, and nonvitrified dung. In summary, high classification accuracies achieved when using field spectroscopy data prove that remote sensing techniques can be used to exploit the spectral differences among the abovementioned soil types in mapping archaeological feature characteristics of farming communities’ settlements.</jats:p>

Island-wide characterization of agricultural production challenges the demographic collapse hypothesis for Rapa Nui (Easter Island)

“…Many instances remain where detecting archaeological landscapes is difficult using traditional multispectral datasets (VNIR). Detecting subtle changes to vegetative health, moisture retention properties of soils and plants, and even differentiating between soils and soil composition are crucial ( 39 – 43 ). This case study demonstrates that SWIR data can distinguish archaeological landscape components that often blend in with their surroundings (like rock gardens) from their surroundings where other components of the electromagnetic spectrum do not have enough spectral discriminatory power to make accurate identifications.…”

Integrating spatial and legacy data to understand archaeological sites in their landscape. A case study from Unguja Ukuu, Zanzibar

“…Airborne remote sensing (multi‐spectral imagery, Synthetic Aperture Radar [SAR] and Light Detection and Ranging [LiDAR]) alongside survey, near‐surface geophysics, and GIS analysis are increasingly used by archaeologists across the continent (Klehm & Gokee, 2020), often in combination with ground‐based techniques (Thabeng et al, 2020). Remote sensing for the identification of settlement distribution patterns based on aerial and satellite photos is common in open (vegetation) landscapes, especially in North Africa (Biagetti et al, 2017; Parcak et al, 2017) and southern Africa (Davis & Douglass, 2020; Sadr, 2016; Thabeng et al, 2019). In densely vegetated and canopied tropical environments, techniques capable of penetrating dense canopies, such as LiDAR, have been most transformative.…”

Geospatial Analysis in African Archaeology: Current Theories, Topics, and Methods

“…The availability, affordability, and reliability of imagery such as Landsat, ASTER, SRTM, and Google Earth showcase how widely available imagery can be used effectively across multiple scales and geographic contexts (e.g., Bubenzer et al 2018;Harrower and D'Andrea 2014;Khalaf and Insoll 2019;Sadr and Rodier 2012;Sampson et al 2015). Case studies that use high-resolution satellite imagery (Biagetti et al 2017;Parcak et al 2016;Reid 2016), hyperspectral imagery (Thabeng et al 2019a), and LiDAR imagery (Sadr 2016) have further improved our ability to assess anthropogenic landscapes at increasingly finer scales. Applications range from classifying natural and anthropogenic landscapes (Schmid et al 2008; this issue) to building predictive models (Davis et al 2020;Klehm et al 2019;Thabeng et al 2019b) and further include monitoring vulnerable cultural heritage (Elfadaly et al 2018;Parcak 2017;Parcak et al 2016;Rayne et al 2017; see also the EAMENA Project [https://eamena.arch.ox.ac.uk/]).…”