Satellite Remote Sensing

Satellite remote sensing technologies have triggered improvements in archaeological research and developments of new tools in archaeological prospection from discovery to monitoring, from documentation to preservation of cultural resources. Nevertheless, this increasing interest in remote sensing has not been accompanied by new perspectives in data processing, analysis and interpretation. Specific methodologies must be developed ad hoc for archaeology in order to optimize the extraction and understanding of the information content from the numerous active and passive satellite data sets. This chapter provides a brief overview on qualitative and quantitative data analysis from visual interpretation to digital manipulation.

The goal of image processing for archaeological applications is to enhance spatial patterns and/or local anomalies linked to ancient human activities and traces of palaeo-environments still fossilized in the modern landscape. In order to make the satellite data more meaningful for archaeologists and more exploitable for investigations, reliable data processing may be carried out. Over the years a great variety of digital image enhancement techniques have been devised for specific application fields according to data availability. Nevertheless, only recently these methods have captured great attention also in the field of archaeology for an easier extraction of quantitative information using effective and reliable semiautomatic data processing. The setting up of fully-automatic methodologies is a big challenge to be strategically addressed by research communities in the next years.

The extraction of the huge amount of information stored in the last generation of VHR satellite imagery, is a big challenge to be addressed. At the current state of the art, the available classification techniques are still inadequate for the analysis and classification of VHR data. This issue is much more critical in the field of archaeological applications being that the subtle signals, which generally characterize the archaeological features, cause a decrease in: (i) overall accuracy, (ii) generalization attitude and (iii) robustness. In this paper, we present the methods used up to now for the classification of VHR data in archaeology. It should be considered that: (i) pattern recognition and classification using satellite data is a quite recent research topic in the field of cultural heritage; (ii) early attempts have been mainly focused on monitoring and documentation much more than detection of unknown features. Finally, we discuss the expected improvements needed to fully exploit the increasing amount of VHR satellite data today available also free of charge as in the case of Google Earth.

The application of pan-sharpening techniques to very high resolution (VHR) satellite data can fruitfully improve the enhancement of archaeological marks and facilitate their detection. Nevertheless, the quantitative evaluation of the quality of the fused images is one the most crucial aspects in the context of data fusion. This is because (i) data fusion application is a rather recent technique applied to archaeology; (ii) the criteria generally adopted for the data fusion evaluation can not fit the needs of this type of application. This chapter provides an overview of pan-sharpening techniques and quantitative evaluation of their capability in (i) preserving spectral fidelity and (ii) sharpening spatial and textural content.

It’s a fact that today archaeologists can interact ever more closely with the modern sciences of surveying, which as a whole may fall under the name of Geomatics. A scientific approach to Archaeology may be in fact reflected in the first place in the adoption of methods, both objective and based on rigorous criteria and tools, for knowing and describing the goods. Remote Sensing, as a technique for surveying a site from a distance by means of special detectors and using management methods based on mathematical and physical models, is by nature a practical expression of this phenomenon, whether it is meant in the most common meaning of use of satellite data or extending its acceptation to aerial photogrammetry. The paper will review some of the issues related to the use of Remote Sensing in the archaeological context, but particularly in relation to other disciplines of Geomatics; Remote Sensing is not, as noted, a technique that is “self contained” and restricted in its scientific and technical areas independently, but it relies on the wealth of other techniques and scientific fields, and they must fit perfectly, especially for applications in Archaeology which have by nature a highly multidisciplinary vocation.

The chapter concerns the research activity conducted by the Institute for Archaeological and Monumental Heritage of the Italian National Research Council in the ancient city and in the territory of Hierapolis in Phrygia (Pamukkale, south-west Turkey), in cooperation with the Italian Archaeological Mission. The research, conducted during the 2003–2010 campaigns, aimed to the reconstruction of the layout of the city and to the study of the ancient topography of its territory. It is based on the integration of different non-destructive study methods and technologies: systematic archaeological and topographical surveys, geophysical prospecting and processing, analysis and interpretation of aerial photos and multi-temporal high resolution satellite images (in particular, space photos of 1960s and 1970s taken by USA reconnaissance satellites; panchromatic and multispectral images taken by QuickBird-2 and Ikonos-2 between 2002 and 2009). During the research, satellite images were not only used to replace and integrate the aerial photographs in the finding and location of traces of archaeological and paleo-environmental elements, but they were accurately corrected in their geometric distortions and employed to integrate available cartography, as base-maps for the field works and in the GIS and webGIS of Hierapolis and its territory.

Although not its primary mission, NASA’s remote sensing missions have been providing archaeologists with useful and unique data since the launch of the very first earth observing missions. While imaging the Earth with a variety of multispectral and hyperspectral instrument mounted on both orbital and suborbital platforms, NASA scientists and collaborators from international universities have discovered, delineated and analyzed archaeological sites worldwide using remotely sensed digital imagery. Several of the ten NASA centers have collaborated with archaeologists to refine and validate the use of active and passive remote sensing for archeological use. The Stennis Space Center (SSC), located in Mississippi USA has been the NASA leader in archeological research. Together with colleagues from Goddard Space Flight Center (GSFC), Marshall Space Flight Center (MSFC), and the Jet Propulsion Laboratory (JPL), SSC scientists have provided the archaeological community with useful images and sophisticated processing that have pushed the technological frontiers of archaeological research and applications. Successful projects include identifying prehistoric roads in Chaco canyon, mapping the Lewis and Clark Corps of Discovery route, assessing prehistoric settlement patterns in southeast Louisiana, discovering the lost city of Ubar and mapping ancient irrigation channels at Angkor. The Scientific Data Purchase (SDP) managed from SSC solicited commercial remote sensing companies to collect archaeological data. Recently, NASA formally solicited “space archaeology” proposals through its Earth Science Directorate and continues to assist archaeologists and cultural resource managers in doing their work more efficiently and effectively. Hyperspectral data offers new opportunities for future archeological discoveries.

Illegal excavations represent one of the main risk factors which affect the archaeological heritage all over the world, in particular in those countries, from Southern America to Middle East, where the surveillance on site is little effective and time consuming and the aerial surveillance is non practicable due to military or political restrictions. In such contexts satellite remote sensing offers a suitable chance to monitor this phenomenon. The chapter deals with the results obtained on some areas of Cahuachi (Peru) by using a time series of very high resolution satellite images. The rate of success in detecting changes related to archaeological looting has been fruitfully improved by adopting a semiautomatic approach based on spatial autocorrelation.

For more than a decade the multinational (Australian, French, Cambodian) Greater Angkor Project has been investigating the rise and fall of medieval urbanism at Angkor, in Cambodia, using a diverse range of techniques, including extensive use of remotely sensed imagery to find, map and analyse elements of urban form. The research activities have focussed on the role of Angkor’s elaborate water management system in the demise of the urban complex, and has recently been expanded to include nearby ‘secondary’ settlement complexes such as provincial centres and ephemeral capitals. In such a research agenda, it is crucial to gain a full understanding of the original hydrological layout of the Angkor basin, in order to provide essential insights into human modifications to the natural hydrology and topography. To this end, a number of multispectral satellite images (including QuickBird and ASTER) were processed and analysed to identify palaeo-environmental traces and anthropogenic features relevant to the identification of remnants of the original fluvial system. Vegetation indices (VI), Vegetation suppression and Principal Component Analysis (PCA) were adopted as the primary procedures in order to detect relevant traces over differing environments such as perennially forested zones, scrubland and barren terrain. The outcome of this work has been to add significant chronological resolution to the current map of Greater Angkor.

Archaeological research into the ancient Jabali silver mines in northern Yemen is a multidisciplinary project linking archaeologists, historians, and geologists, supported by remote satellite-based sensing. Mining companies are considering the development of the Jabali deposit for zinc ore. Some of the facilities, including a large open pit, will have a destructive impact on the old pits. Multispectral (Landsat TM, Terra ASTER, ALOS-AVNIR-2, etc.) and very high spatial resolution images such as QuickBird have been widely used for the geological setting and detail mapping of the different archaeological sites. The remote sensing data also provides a solid basis when it comes to detecting current operational sites.

Sri Ksetra is a very large urban site created by the Pyu people, ca. fourth century AD, on the southern edge of the Dry Zone of Central Burma. Ancient irrigation works, discovered by Stargardt in 1985–1988 and first published in 1990, were essential to its urban development and, though rarely visible as surface features today, still profoundly affect the post-destruction settlements on the site and their rice harvests. This paper documents the sub-surface and surface movement of moisture across the site after the end of the monsoon season, i.e. in the period from September to February. It is based on the study of combined resources, including spaceborne multi-spectral and multi-temporal satellite imagery, aerial photographs and archaeological maps. It reveals how surface and sub-surface water still flows along the ancient irrigation channels and collection tanks of the site and demonstrates that an ancient system, which has been in an advanced state of sedimentation for over 1,000 years, still functions mainly at a subsurface level, creating differences in soil productivity and drainage today. The study explores the potential value of freely available imagery of medium resolution for archaeological research on other ancient landscapes embedded in modern ones where differences of wetness and vegetation are significant. Finally, these satellite images record the presence of ancient ritual ponds and associated burial terraces in the southern extramural area in a degree of detail that, surprisingly, compares well with the aerial photographs taken over 50 years ago when surface disturbance was at a lower level.

Precious information to reconstruct ancient environmental changes, still fossilized in the present landscape, may be captured from multispectral satellite images from medium to high spatial resolution. In particular, satellite derived moisture content may facilitate the identification of areas involved in early environmental manipulation mainly addressed to set up irrigation and artificial wet agro-ecosystems where the natural rainfall was insufficient to support agriculture. Up to now, only a few number of archaeological studies on spatial patterns of moisture have been carried out through the world using satellite optical data. In this chapter, Landsat and ASTER data were analyzed for some areas near Nasca river within the drainage basin of the Rio Grande, densely settled over the centuries and millennia even if the physical environment presented serious obstacles to human occupation. This region is one of the most arid areas of the world, so that the pluvial precipitations are so scarce that they can not be measured. To face this critical and extreme environmental conditions, ancient populations of the Nasca River valley, devised an underground aqueducts called puquios, some of which are still used today. Archaeologists suggest that during the Nasca flourishing period, certainly the number and spatial distribution of puquios was larger than today. We used satellite data to identify areas to be further investigated to assess if and where therein puquios were constructed for water control and retrieval.

Traditionally, archaeological aerial surveyors fly routes selected on the probability of site visibility and record only what they perceive as significant. This approach to aerial reconnaissance has come under criticism in recent years with respect to productivity, bias and cost. Through panchromatic imagery with sub-metre resolution and high-resolution multispectral products, modern commercial satellites are capable of recording fine details of both extant sites and archaeological cropmarks. They are not limited by fuel costs, airspace laws, pilot time or conflicting regions of interest. In addition, the multiple wavelength images and powerful image analysis software available offer the facility to manipulate images in order to maximise the visibility of surface details and to detect any variations in the non-visible spectrum. The present paper employs a comparative analysis of above-ground remotely-sensed data recovered on various dates and with various methods of acquisition (QuickBird and Corona satellite imagery; WWII vertical aerial photography and recent oblique aerial photographs), in order to discuss the potential of satellite imagery to reveal buried archaeological features in the area of southern Dobrogea (Romania) and to assist in the reconstruction of past landscapes. It outlines some of the methodological approaches employed, highlights areas for further research and presents some preliminary outcomes.