Tiree: Aerial and Satellite Archaeological Prospecting
Multi-spectral aerial photography, LiDAR survey, and desk-
based assessment
Kelly Kilpatrick and Rob O’Hara – Archer Heritage Ltd
NL 97300 44940 06 Create and analyse remote sensing datasets
(LiDAR, thermal, satellite, raster vegetation indices) of the island of Tiree. The survey took place on 6 May 2021.
An objective of this project was to generate multiple remote
sensing surveys of the island of Tiree for archaeological
prospection. Remote sensing technologies offers a nondestructive
approach to identify study and preserve specific sites and entire
archaeological landscapes. Tiree was chosen for examination
because sandy machair habitats can work well with remote
sensing methods; multiple existing remote sensing datasets
were freely available; and the island has a long history of human
habitation. Early-medieval documentary sources indicate the
presence of several early Christian monasteries on the island but
their locations are now unknown. One aim of this project was to
test the feasibility of remote sensing to identify early Christian
ecclesiastical foundations.
This project acquired the 2002 airborne data for Tiree and Coll
collected for a report for Historic Scotland published in 2003 and
LiDAR of the island made by Scottish Natural Heritage in 2007.
Multispectral satellite imagery covering the SW of the island taken
by the Pleiades P1B satellite in September 2018 was purchased
from Land Info Worldwide Mapping.
The LiDAR data generated six visualisations (slope, hillshade,
multidirectional hillshade, principal component analysis of
multidirectional hillshade, local relief models, and sky view). The
satellite data was supplied with basic radiometric processing and
additional processing included top of atmosphere reflectance and
cloud masking. Because of the close correlation between vegetation
health and buried archaeological features multispectral imaging
has been widely used for archaeological surveying. The processed
and calibrated satellite image was transformed into two raster
vegetation indices. A ratio was calculated using near infrared
and red bands to create a Normalized Difference Vegetation Index
(NDVI). Local NDVI anomalies were highlighted by determining
mean NDVI intensity within a set radius of 5m and 10m then
subtracting this from NDVI pixel values The LiDAR thermal and
satellite datasets are visualized in ArcGIS 10.5.
All digital data created in this project will be archived with
Historic Environment Scotland and made Open Access. Analyses
of the remote sensing survey of Tiree were still ongoing as this
summary was being prepared.
Archive: NRHE
Funder: Society of Antiquaries of Scotland
(Discovery and Excavation in Scotland (2021), p. 46)