You must have a javascript-enabled browser and javacript and stylesheets must be enabled to use some of the functions on this site.


Potential of ALOS PALSAR Imaging for Investigation of Archaeological Underground Marks

Nicole Dore(1)

(1) University of Rome, P.le Aldo Moro, 5, 00185, Italy


Since June 2006 ALOS-PALSAR has been acquiring L-band Synthetic Aperture Radar (SAR) images of Earth surface from space. Fully polarimetric L-band data properties have been exploited in many land applications, including discrimination of different kinds of surfaces and objects. Analysis of scattering matrices at L-band represents an important step towards the retrieval of bio-geophysical parameters.

This contribution reports on a study we carried out to understand the L-band response over a variety of different archaeological sites on the Earth’s surface and underground.

Since the early 1900’s, after pioneer attempts by balloon, aerial photography was used mainly for military purposes as RAF (Royal Air Force) acquisition flights. This was developed greatly during the Second World War. Only subsequently has this kind of airborne imagery been used for scientific purposes. Scientific applications include archaeology, where airborne imagery has been use to locate and investigate archaeological sites all over Europe. Archaeological marks singled out from aerial photographs have been confirmed by excavations subsequently carried out in situ.

Since the beginning of 2000, optical satellite imagery has provided an innovative contribution to archaeological remote sensing, using IKONOS (1 m using PAN and 4 m using MS) and QUICKBIRD 2 (0.61 m at nadir using PAN and 2.4 m at nadir using MS).

The use of SAR imagery is very innovative for archaeological research. Despite the capability of optical instruments to detect soil marks, these sensors cannot see under the ground, or through vegetation and clouds. Radar acquisition, instead, can provide information day and night and all through the year, giving additional information concerning electromagnetic properties of the target, especially when, for each pixel, all elements of the scattering matrix are available.

We selected test sites located in Italy: Veio, Vulci and Cerveteri are etruscan-roman sites in the north of Rome; Arpi is a “dauno” site in southern Italy near Foggia. These areas are archaeologically well known, thanks to aerial photographs, the data of which has made possible the photogrammetric restitution of the sites.

We have considered alternating-pol images from ENVISAT-ASAR and dual- and quad-pol images from ALOS-PALSAR. Our choice is linked to their potential to discriminate the backscatter response as a function of the morphological features of each site. Optical data sets have been used as a support to the study. Several sensors have demonstrated their utility: ALOS-PRISM, AVHNIR 2, SPOT-2, SPOT-4, KOMPSAT-1, KOMPSAT-2 and IKONOS.

Polarimetric SAR sensors are expected to support the detection of underground archaeological remains by observing soil anomalies in terms of moisture content and vegetation growth. In particular, estimates of moisture content may be useful to detect potential archaeological remains that cause moisture stagnation around them. Archaeological marks are also detectable because of underground structures that hinder regular growth of vegetation roots. For this reason, we have selected images from different periods in the year, when vegetation goes through different levels of development.

An initial limitation of the method is tied to the resolution of PALSAR (min 10 m/max 20 m), which enables the recognition of large areas, but for single infrastructural elements a higher resolution is needed, since the researched archaeological structures are typically smaller than 10 m. Finally our work will be supported and integrated by archaeological research in situ and ancillary information in order to have a complete case history that can be used for the detection of other archaeological sites and increase the synergy between humanistic and scientific knowledge.


Full paper


  Higher level                 Last modified: 07.05.06