High Resolution Satellite Imagery and archaeology: a theoretical approach [1]

D. Danelli

University of Milan (Italy)

To the present day, the use of Earth Observation satellites is widespread and covers many fields of human expertise, such as cartography, agriculture, forestry management, and climate control. Even in archaeology, satellite imagery has progressively become a powerful asset, in two different ways: it can contribute to the management and the safeguard of known cultural heritage, and it can be used as a tool to detect new potential archaeological contexts, a role shared (and, more importantly, integrated) with more classical methods of Remote ensing.

This brief paper deals with what is nowadays called “High Resolution Satellite Imagery”, but this is, in my opinion, a third phase in the History of Satellite Imagery. In fact, it is preceded by a phase of experimentation, in which the first images from the Earth orbit are taken and subsequently used for a scientific purpose, and a phase of declassification, in which previously top-secret imagery is made available to common people.

In order to fully understand the capabilities of contemporary satellites, and the potentialities related to a daily high-resolution coverage of the Earth surface from the orbit, an overview of the first two phases seems necessary.

1. Experimentation. The first images taken from outside the stratosphere predate the launch of the first satellite. In fact, they were taken shortly after World War II: on the 24th October 1946, a US team of scientists at the White Sands missile base (New Mexico) launched a V2 missile, confiscated from Germany after the War. Encased in the missile there was a small camera, set to take a picture every 1,5 seconds. After the launch and the subsequent crash, only the metal case that protected the impressed film survived, and this allowed us to see for the first time the Earth surface from a distance of 105 km (before that date, the highest altitude from which a picture was taken was 22 km, in 1935) [Reichhard, 2006. A video service about this event can be seen here]. Of course, a missile is not a satellite; nonetheless this event represents a milestone, and the beginning for a new era of exploration (fig. 1).

After eleven years (and more than a thousand images taken from later V2 launches, until 1950), the award for being the first nation capable to take to orbit a functioning satellite goes to the USSR: in fact, from 4th to 26th October 1957 Sputnik 1 sends its radio pulses to the surface, triggering the so-called Space Race. Unfortunately, given the unstable political situation of that period, efforts in creating satellites capable of taking images of the Earth surface were mainly pointed towards espionage, by both the most advanced world powers. In fact, behind the Discoverer scientific satellites launched by the USA, there were KeyHole espionage missions; USSR did the same with some of the Kosmos missions that concealed, behind the generic name of the Soviet’s space mission, top-secret Zenit espionage satellites. The role of these satellites for archaeology will be discussed later.

It was not until 1972 that NASA launched the first scientific Earth Observation satellite: its name was Earth Resources Technology Satellite (ERTS–1), later renamed LANDSAT–1. It was equipped with three Return Beam Vidicon cameras and a 4–band MultiSpectral Scanner, but it could not reach more than 80m of Ground Sampling Distance: obviously, it could not compete with classical aerial photography in matter of detail level. Imagery taken from LANDSAT was mainly used in geography and cartography, to characterize environments, landscapes or agricultural patterns in vast portions of the Earth surface, or to create small scale maps in previously uncharted areas, such as Near East countries, Central Asia, Africa, and Central America. Despite this, the potential of satellite imagery for archaeological practice had been already recognized five years before the launch of the ERTS [Thompson, 1967], and in the 1970s and the 1980s many archaeologists used Landsat’s MultiSpectral Scanner to investigate ancient landscapes [Quann, Bevan, 1977; Adams, 1981; Richards, 1989]. After LANDSAT, imaging payloads equipped on satellites kept improving, leading to some noticeable archaeological result [2]. However, everything changed with the 1990s and the dissolution of the Soviet Union.

2. Declassification. After 1991, the less tense situation between the Eastern and Western countries led to the progressive declassification of photographic material obtained by spy satellites. The first country to remove the top-secret status from this imagery and commercialise it is the Russian Federation: in 1992, a first set of images with a resolution of the order of 2m was released by Sovinformsputnik, and in 1998 a second set, with a resolution of 1m, followed [Commercial Satellite Imagery, 2002]. The first set had a significant impact on the archaeological world: until then, the maximum level of detail obtainable from satellite imagery was approximately 10m, from France’s SPOT satellite. The 2m resolution of the Russian dataset, instead, allowed a much more complete vision of the landscape, including both upstanding archaeological features and large crop- and soilmarks. Archaeological work that uses declassified Russian imagery is mainly focused on the areas of known archaeological importance, such as Stonehenge [Fowler, 1996; 2001] and Zeugma on the Euphrates [Comfort, 1997; Comfort et al., 2000].

The second Russian declassification, on the other hand, had less resonance in archaeology. In fact in 1995, three years before it, another event occurred, that dramatically changed the impact of satellite imagery for archaeological practice: the first USA declassification. With the Executive Order 12951 of the 22nd February 1995, President Clinton removed the secrecy clause from 860.000 photos taken from Corona (codename KH–1 to KH–4), Argon (KH–5) and Lanyard (KH–6) missions, from 1960 to 1972, and copies of them were made available for purchase from the US Geological Survey. These images almost have a full coverage of the Earth surface, and, in the most recent photos (KH–4B), spatial resolution could almost reach 0,50m. Subsequently, a new declassification in 2002 granted access to 18.000 photos from KH–7 Gambit missions (1963–1967) and 29.000 from KH–9 Hexagon missions (1973–1980). Despite the better spatial resolution of Gambit imagery, this series of satellites was mainly focused on military targets, whereas KH–9 images declassified in 2002 are from the lateral cameras of the optical payload. Pictures taken from 1971 to 1986 with the primary camera of KH–9 satellites have been released in 2011, but, to this day, the two million images are only available at NARA, the Central National Archive of the United States. So far, pictures taken from espionage satellites that have proven themselves to be more useful for archaeological purpose are those from KH–4A and KH–4B mission: the almost global coverage, coupled with a spatial resolution comparable to some modern era satellites, resulted in more than 50 archaeological papers, dating from 1998 to 2016 and primarily listed in two significant articles by Martin J. Fowler [Fowler, 2004a; 2013] (Table 1).

Table 1.
Archaeological projects based on declassified satellite imagery (1998–2016)

Place Mission Archaeological Use Bibliography

Turkey 1103, 1104 First ever overhead photograph of the Roman city of Samosata on the Euphrates.
Mapping, interpretation and illustration of key places, features, and geomorphology in the region of the Roman city of Zeugma prior to flooding of the area following the construction of the Birecik dam. [Kennedy, 1998a; 1998b]

England 1104 Investigation of the archaeological potential of KH–4B imagery in Hampshire. [Fowler, 1997]

Senegal Corona
Argon Use of Corona and Argon imagery to assess environmental changes in Senegal between 1960s and 2000. [Tappan et al., 2000]

Syria Not specified Aided geomorphological study of the complex landscape hinterland of the Bronze Age site of Tell Brak. [Wilkinson et al., 2001]

Syria 1102, 1108, 1117 Identification of possible archaeological sites in the hinterland of Tell Hamoukar. The photographs were found to be extremely powerful for site identification since they preserve a landscape that is over thirty years old, prior to the expansion of towns, the intensification and mechanisation of agriculture and the introduction of diesel pumps and irrigated cotton fields. [Ur, 2002]

Syria 1102, 1105, 1108 Identification and mapping of ancient road systems dating from the early Bronze Age in the Upper Khabur basin in north eastern Syria, where conventional aerial photography is limited. [Ur, 2003]

Syria 1108, 1110, 1111 Aided identification of non-Tell settlement remains and distribution and organisation of field systems in Homs region. [Philip et al., 2002a; 2002b;
Donoghue et al., 2002]

Syria 1038 Provided broader landscape survey as part of an interdisciplinary investigation of an eighth to twelfth century Islamic industrial complex associated with the city of al-Raqqa in northern Syria. Predating both the modern urban expansion of Raqqa and development of extensive irrigation systems in the hinterlands, Corona photographs have proved invaluable for mapping the now largely destroyed cultural landscape. [Henderson et al., 2002]

Iran Not specified Used together with Spot panchromatic imagery to illustrate specified landscape change on the Susiana plain in southwestern Iran between the late 1960s and early 1990s. During this period, much of the plain was levelled and an extensive network of irrigation channels constructed, which now mask traces of earlier irrigation systems of Sassanian and Islamic date that can be seen as dark lines on Corona photographs [Kouchoukos, 2001]

Jordan 1115 Used to compare the appearance of features associated with Rome’s desert frontier visible on Corona images with conventional aerial photographs of the sites. [Fowler, 2004b]

Ukraine Not specified Part of a multitemporal, multiresolution database of satellite imagery of Chersonesos, Crimea, an area that is one of the best preserved examples of an ancient Greek agricultural landscape. The site’s contemporary strategic position limits physical access and thus remote sensing offers an effective and reliable means of mapping the ancient landscape as well as monitoring urban encroachment over the past thirty years. [Trelogan et al., 2002]

Turkmenistan Not specified Used as a dataset within a GIS to investigate settlement distribution relating to evolution of the ancient irrigation system of the river fan of the Murghab delta before the modern building of the Kara Kum canal, completed in 1988. [Cerasetti, Mauri, 2002]

Armenia 1115 Used together with unsophisticated computer hardware and software to identify features including cultivation terraces, stone clearance cairns, enclosures, and sites of shepherds’ camps. [Palmer, 2002]

Romania 1022 Used in place of large scale background maps for rectification and transcription of conventional oblique aerial photographs covering the Roman site of Micia. [Oltean, 2002]

Near East Not specified Summary monograph on the archaeological landscapes of the Ancient Near East, which extensively exploits the Corona images. [Wilkinson, 2003]

Syria Not specified KH–4B photographs were used to investigate the topography and cultural landscape of the early Islamic city of al-Raqqa in northern Syria. [Challis et al., 2004;
Challis, 2007]

Russian Federation 1042 KH–4A photographs were used as part of an archaeological survey of Bronze Age, Iron Age and ancient Turks monuments in the Altai Mountains of South Siberia [Gheyle et al., 2004;
Goossens et al., 2006]

Armenia 1115 KH–4B photographs were used together with digital photographs taken at low altitude from a paramotor were used to investigate a prehistoric, Medieval, and Soviet-era landscape in Armenia. [Faustmann, Palmer, 2005]

Iraq 1102, 1108, 1117, 4017, 4031 KH–4B and KH–7 photographs were used to describe canals and other archaeological features related to water management and settlement in the landscape in the vicinity of the Assyrian city of Nineveh. [Wilkinson et al., 2005;
Ur, 2005]

England 1104 Ring ditches and a possible oval enclosure were detected as cropmarks on a KH–4B photograph acquired in August 1968. [Fowler, Fowler, 2005]

Iraq Not specified Corona satellite photographs were used to supplement aerial photography to produce a photogrammetric plan of the ancient city of Samarra on the east bank of the Tigris. [Northedge, 2005]

Turkey 1112 Used KH–4B photographs together with archaeological survey data to investigate aspects the settlement history of the Amuq Valley in the northern Levant including the archaeological landscape of Late Roman Antioch. [Casana, 2004; 2007]

Iran 1103, 1110 Mapped using KH–4B photographs pastoral and irrigation landscapes in north-western Iran that have been virtually obliterated since the photographs were acquired. [Alizadeh, Ur, 2007]

Iran 1052 KH–4A photographs were used to map the natural landscape of the Bushehr peninsula of Iran and to highlight changes to the landscape caused by modern water extraction and agriculture and to map individual monuments within the landscape. [Challis, 2007]

Syria 1108, 1110 Archaeological sites captured on KH–4B were accurately located using a recent Ikonos satellite image of the Homs Region of western Syria and provides advice in the design and costing of surveys using satellite imagery. [Wilkinson et al., 2006;
Beck et al., 2007]

Egypt 1049 KH–4A photographs that revealed surface features that have been obscured by recent agricultural developments, together with field investigations of Holocene sedimentary deposits, were used to create a palaeogeographic map that places Late Bronze Age archaeological sites in Northwest Sinai in their environmental context. [Moshier, El-Kalani, 2008]

Russian Federation 4028 A KH–7 photograph was used to describe the material culture of the Cold War installations around Moscow that were associated with the former Soviet Union’s first generation of surface to air missile (SAM) system. [Fowler, 2008]

Turkey, Syria 1105 The stereo capabilities of KH–4B photographs were utilised to produce high-resolution digital elevation data and maps of archaeological landscapes in the Levant. [Casana, Cothren, 2008]

Turkey Not specified Orthorectified KH–4A imagery was combined with LANDSAT Thematic Mapper, QuickBird and ASTER imagery to investigate the archaeological site at Tilmen Höyük (south-eastern Turkey). [Bitelli, Girelli, 2009]

Syria 1046, 1105 The Tell Tuqan Survey Project used multi-temporal KH–4A and KH–4 photographs together with QuickBird imagery to study the tell and the surrounding area to inform the reconstruction of the ancient topography of the region. [Castrianni et al., 2010]

Italy, Turkey 9022, 1043, 1049, 1103, 1107, 1109, 1111, 4036, 1210, 1216. KH–3, KH–4A, KH–4B, KH–7 and KH–9 photographs were used to illustrate the importance of historical aerial and satellite photographs support archaeological and geo-archaeological research in Italy and Turkey. [Scardozzi, 2010]

Iran 1035, 1045 The evolution of the Lower Khuzestan plain in SW Iran was investigated using KH–4A photographs, LANDSAT imagery together with archaeological, geological, and historical datasets. [Walstra et al., 2010]

Sudan Not specified KH–4B and QuickBird imagery to support archaeological research and Cultural Resource Management in the Sudanese Middle Nile. [Edwards, 2010]

Latvia 4031, 1110, 1210 KH–7, KH–4B and KH–9 photographs were used to describe the development of the Hen House ballistic missile early warning radars at Skrunda in Latvia between 1966 and 1975. [Fowler, 2010]

Syria 1021, 1102, 1105 1108, 1117 Over 6.000 km of premodern trackways in the vicinity of Tell Hamoukar, one of the largest Bronze Age sites in northern Mesopotamia, were identified and mapped from KH–4A and KH–4B photographs. [Ur, 2010]

Israel 1111, 1115 A comparison of two KH–4B photographs of the Roman siege-works at the fortress of Masada shows the impact of the solar lighting conditions at different times of the day on the appearance of the landscape and upstanding archaeological feature. [Fowler, 2011]

Iraq KH–4B e KH–7 Used KH–7, KH–4B, QuickBird–2, Ikonos–2 and WorldView–1 to study the ancient city of Ur, inaccessible since 2003. [Di Giacomo, Scardozzi, 2012]

Near East Not specified. Declassified satellite imagery was used to study environmental and climatic changes, and to map new archaeological discoveries obtained through satellites in vast portions of the Middle East. [Casana et al., 2012]

Syria Not specified Use of declassified satellite imagery to study places where political choices and/or deep vegetation prevent other ways to acquire aerial data. [Hritz, 2013]

China 1101, 1115 Study about potentiality of Corona imagery on Chinese soil, focusing in particular on the ancient capital city of Qufu and on the post-war agricultural development. [Min, 2013]

Armenia 1115 Use of KH–4B imagery taken in Armenia on the 20th September 1971 to research potential archaeological evidence. [Palmer, 2013]

India 1117 Use of KH–4B imagery to survey monsoon floodplains in northern Gujarat and to analyse the evolution of agricultural and industrial landscape. [Conesa et al., 2015]

Russian Federation Not specified On a KH–9 picture of the Kapustin Yar missile base East of Volgograd it was possible to see a landscape graffiti depicting the name ЛЕНИH . [Fowler, 2016a]

USA 1217-4 Analysis of landscape and archaeological features of a picture taken from the primary camera of a KH–9 Hexagon satellite representing the city of St. Augustine in Florida, where the Spanish fort Castillo de San Marcus was built between 1672 and 1695. [Fowler, 2016b]

Back
Regarding these data, it is possible to make some observations. First of all, the papers considered are 52, concerning 21 different states: 12 in Asia (Turkey, Syria, Iran, Jordan, Turkmenistan, Armenia, Altaj Republic, Iraq, Israel, China, India, and the Sinai Peninsula), six in Europe (England, Ukraine, Romania, Russian Federation, Latvia, and Italy), two in Africa (Sudan, Senegal) and one in the US (Florida). A major interest in Asian, and specially Near and Middle East countries, is discernible if we look to the number of papers: 14 of them deal with Syria, six with Turkey (and one considers them both), four each with Iran and Iraq, three with Armenia, two with the Altaj Republic, two with Near East altogether, and then one each with Jordan, Israel, the Sinai Peninsula, Turkmenistan, China, and India, totalling 42 out of 52 papers (other countries considered are Russian Federation and England, with two papers, and one each for Ukraine, Romania, Latvia, USA, Senegal, and Sudan [3]) (fig. 2). As for the missions, it can be seen that, in the 43 papers where the number of the mission is noted, there is a majority of Corona KH–4B satellites (mission 1101 to 1117), since it had the best optical payload equipped in all the Corona series: 59 out of the 84 satellite missions whose imagery was used in these papers are from KH–4B missions; 13 from KH–4A and one from previous Corona missions (9022, from KH–3). Adding all data from Corona missions, it turns out that 73 out of 84 missions are from Corona satellites, while only six refer to Gambit KH–7 missions (n. 4001 to 4038) and five to the Hexagon KH–9 project (n. 1201 to 1220): this is consistent with the higher archaeological usefulness of Corona imagery (in the 43 papers where it is said which missions are being used, 40 of them refer at least to imagery from one Corona mission).

As was foreseeable, the major focus of these studies is in Asia, from Near and Middle East to Central Asia (Turkmenistan and Altaj Republic), for three main reasons: there is a higher quantity of KeyHole imagery regarding Eastern countries, and they all have a really high intrinsic archaeological interest. This, added to the lack of previous cartographic and aerophotographic data, resulted in a major interest from archaeologists: declassified satellite imagery from the 1960s in Middle East plays the same role that vertical aerial photography had played for Europe twenty years before: it is used as a systematic aerial coverage for photointerpretation, and it depicts the human landscape before the major changes caused by massive industrialization processes, when the landscapes were more similar to historical ones than today’s.

In this sense, declassified imagery still has an undeniable archaeological role in all those studies that imply a historical approach to satellite imagery, or deal with history and evolution of the human landscape. However, a new era for satellite imagery begins with the introduction of commercial satellites: these were not designed for espionage or scientific purpose, but to sell the best possible images for potential buyers. This, combined with a constantly improving technological level, brought to a huge leap during the last twenty years.

3. High Resolution (HR) Satellite Imagery. On the 24th of December 1997, a company called EarthWatch launched its first commercial satellite, EarlyBird–1. It was the first satellite built with the only purpose of taking high resolution Panchromatic (3m) and MultiSpectral (15m) imagery of the Earth surface from the orbit, and to sell them to anyone who could afford it. Unfortunately, EarlyBird–1 lost every contact four days after the launch, and no image was ever taken. EarthWatch knew that high resolution satellite imagery could be a very profitable market, and two years later, on the 24th of September 1999, managed to get in orbit a second satellite, Ikonos. It had amazing capabilities: at nadir, it could reach a Ground Sampling Distance of 0,81m in Panchromatic (PAN) mode and of 3,2m in four MultiSpectral (MS) bands (blue, green, red, near infrared – NIR). Ikonos became the first HR satellite to sell its imagery, and one of the most used until it stayed in orbit (31st March 2015). EarthWatch, the company that owned it, in 2001 changed its name and became DigitalGlobe, and is now the world leader of High Resolution satellites.

However, quantifying Scientific and Commercial satellites with optical payloads launched since 1997 has proven itself to be almost impossible. After a long research, that involved all commercial satellite companies and every research institution that has managed to achieve successful launches, 49 satellites have been selected, according to their resolution; in the following diagram (fig. 3), they are listed by date of launch, and, where available, details about optical and MS capabilities are given (Table 2).

Table 2.
List of commercial High Resolution satellites (HRS) updated to August 2017

Name Producer Launch date In orbit PAN (m) MS (m)
Notes

EarlyBird–1 EarthWatch 24.12.1997 - 3 15
EarlyBird–1 was the first satellite ever to be built with the sole goal of achieving high resolution images for commercial purpose. Unfortunately, all contacts were lost 4 days after launch, and no image was taken.

Ikonos EarthWatch 24.09.1999 - 0,81 3,2
Originally named Ikonos–2, it was re-named after the launch failure of Ikonos–1 on April 27th, 1999.
In orbit until March 31st, 2015.

EROS A ISI 05.12.2000 ✔ 1,5 /
First Israeli commercial satellite, produced by Israeli Aerospace Industry.

OrbView–4 GeoEye 21.09.2001 - 1 4
Launch failure. First satellite ever equipped with hyperspectral scanners. Ground resolution of OV–1 & –2 was more than 1km, and OV–3 was launched two years later than OV–4.
DigitalGlobe bought GeoEye in January 2013.

QuickBird–2 DigitalGlobe 18.10.2001 - 0,61 2,44
EarthWatch inc. became DigitalGlobe inc. in September 2001. QB–1's launch was not successful. QB–2 stayed in orbit until January 24th, 2015.

SPOT–5 CNES 03.05.2002 - 2,5 10
Fifth Satellite Pour l'Observation de la Terre of the French Centre National d'Études Spatiales. His older brother, SPOT–4, in orbit from 1998 to 2013, had a 10m PAN and 20m MS ground resolution.
SPOT–5 stopped working on March 31st, 2015.

OrbView–3 GeoEye 26.06.2003 - 1 4
No image was taken after April 23rd, 2007. On March 13th, 2011, the satellite was taken down.

FORMOSAT–2 NSPO 21.05.2004 - 2 8
First HRS from Republic of China (Taiwan)'s National Space Organization. Its scanners could achieve 2m PAN images and 8m 4-bands multispectral images. After twelve years, it was decommissioned on 1st Aug 2016.

CartoSat–1
(IRS–P5) ISRO 05.05.2005 ✔ 2,5 /
First HRS produced by ISRO, India Space Research Organisation. It was equipped with a panchromatic camera only.

EROS B ISI 25.04.2006 ✔ 0,6 /
Second HR satellite launched by Israeli Space Agency.

Resurs–DK–1 NTs OMZ 15.06.2006 - 1 2–3
First HR civil satellite of Russian space agency. It remained in orbit longer than expected, and communications finally terminated on February 17th, 2016.

CartoSat-2 ISRO 10.01.2007 ✔ 0,8 /
Second Indian HRS, first to have a better resolution than 1m²/px.

WorldView–1 DigitalGlobe 18.09.2007 ✔ 0,41 /
DigitalGlobe's WorldView, at the moment the constellation of satellites with the highest resolution, was the first to achieve a resolution better than 0,5m. However, until June 2014, only US Government could access the maximum resolution, and denied the commercial use of imagery with a resolution better than 0,5m.

RadarSat–2 CSA 14.12.2007 ✔ 3 /
First HR satellite of the Canadian Space Agency, and first synthetic-aperture radar (SAR) ever to get to the threshold of 3m ground resolution(RadarSat–1's ground resolution was 10x9 meters at its best).

CartoSat–2A ISRO 28.04.2008 ✔ 0,8 /
Third Indian HRS.

GeoEye–1
(OrbView–5) GeoEye 06.09.2008 ✔ 0,41 1,65
The US government prohibition to sell imagery with a resolution better than 0,5m was also in force for GeoEye, later acquired by DigitalGlobe itself.
First MS imagery with ground resolution better than 2m.

WorldView–2 DigitalGlobe 08.10.2009 ✔ 0,41 1,85
First satellite of the WorldView constellation equipped with a MS scanner.

CartoSat–2B ISRO 12.07.2010 ✔ 0,80 /
Fourth HRS of the Indian constellation.

Pléiades–1A CNES 17.12.2011 ✔ 0,5 2,8
First of a two HRS constellation commissioned to the Centre Nationale d'Études Spatiales, SPOT satellites producers, by the ESA, European Space Agency.

KompSat–3 KARI 27.05.2012 ✔ 0,7 2,8
First HRS of the Korea Aerospace Research Institute, South Korean space agency.

SPOT–6 CNES 09.09.2012 ✔ 2 8
Sixth French Satellite Pour l'Observation de la Terre, ten years after the launch of SPOT–5.

Pléiades–1B CNES 02.12.2012 ✔ 0,5 2,8
Second of the two Pléiades satellites.

GaoFen–1 CNSA 26.04.2013 ✔ 2 8
First HRS of the Chinese National Space Agency. Gao Fen in Chinese means high resolution.

Resurs–P1 Roscosmos 25.06.2013 ✔ 1 3–4
Second Russian HRS.

SkySat–1 Skybox Imaging (then TerraBella) 21.11.2013 ✔ 0,9 2
Skybox Imaging, founded in 2009, launched its first HRS back in 2013. Less than a year later (Aug 2014), the company was acquired by Google inc. for half a million USD, and re-named Terra Bella (Italian for Beautiful Earth), from Terra Bella ave., Mountain View, CA, where Google is based.

SPOT–7 CNES 30.06.2014 ✔ 2 8
Seventh French Satellite Pour l'Observation de la Terre .

SkySat–2 Skybox Imaging 08.07.2014 ✔ 0,9 2
Second Skybox/Terra Bella HRS.

WorldView–3 DigitalGlobe 13.08.2014 ✔ 0,31 1,24
The third WorldView satellite, along with the fourth, are equipped with the best optical scanners available on market. Until Feb. 2015 US government forbade the selling of the maximum resolution images.
Its MS scanner can record images in 16 bands (six in visible light, two NIR from 770 to 1040nm, and eight shortwave infrared – SWIR – from 1195 to 2365nm) at a maximum resolution of 1,24m.
On top of that, WorldView–3 is also equipped with a CAVIS scanner (Clouds, Aerosols, Vapers, Ice and Snow) that records images in 12 bands from 405 to 2245nm, with a ground resolution of 30m.

GaoFen–2 CNSA 19.08.2014 ✔ 0,8 3,2
Second Chinese HRS, first with a resolution better than 1m.

Resurs–P2 Roscosmos 26.12.2014 ✔ 1 3–4
Third Russian HRS .

KompSat–3A KARI 25.03.2015 ✔ 0,55 2,2
Second Korean HRS.

GaoFen–8 CNSA 26.06.2015 ✔ ? ?
Probably the best HRS in the Chinese constellation, although technical data have not been released.

TripleSat SSTL 10.07.2015 ✔ 0,8 3,2
HRS triplet with better than 1m resolution launched by Surrey Satellite Technology Ltd., associated with Surrey University.

GaoFen–9 CNSA 14.09.2015 ✔ 0,5 2
Chinese HRS with better known resolution.

Rersus–P3 Roscosmos 13.03.2016 ✔ 1 3–4
Fourth Russian HRS.

CartoSat–2C ISRO 22.06.2016 ✔ 0,65 2
Best Indian HRS at the moment, with a resolution of 0,6m PAN and 2m MS.

SkySat–3 TerraBella 22.06.2016 ✔ 0,9 2
First SkySat launched under Google aegis, on the same missile that carried CartoSat–2C.

PerúSat–1 CONIDA 16.09.2016 ✔ 0,7 2
First satellite of the Comisión Nacional de Investigación y Desarrollo Aeroespacial, Peruvian Space Agency.

SkySat–4/–7 TerraBella 16.09.2016 ✔ 0,9 2
On the same missile that carried PerúSat–1 there were also 4 SkySat satellites of the Terra Bella/Google HR constellation. In the future, said constellation is to be composed of 21 elements.
On 3rd February 2017, Planet Labs inc. stated on its website to have acquired TerraBella from Google.

WorldView–4 DigitalGlobe 26.11.2016 ✔ 0,31 1,24
The launch originally scheduled for June and then for September 2016, was re-scheduled due to the wildfire that burned the woods surrounding the base of Vanderberg, CA, where the launch was set. Along with WorldView–3, it is the satellite which takes the images with the best resolution on the market.

Göktürk–1 Turkish Aerospace Industries 05.12.2016 ✔ 0,8 /
The satellite, commissioned by the Turkish Ministry of Defence to the Italian company Telespazio, was the first Turksih HRS to break the barrier of 1m PAN resolution. His “older brother”, Göktürk–2, was launched on Dec 12th, 2012, and it can take images with a resolution of 2,5m PAN, 10m in four MS bands and 20m in an experimental SWIR band.

GaoJing–1/–2 CNSA 28.12.2016 ✔ 0,5 2
The two GaoJing minisatellites, first steps of a newer and wider Chinese constellation, were launched on Dec 28th, 2016. However, mistakes were made during the launch, and part of the missile payload crashed to the ground. After a couple of weeks of uncertainty about their fate, the two satellites became operative on Jan 12th, 2017.

CartoSat–2D ISRO 15.02.2017 ✔ 0,65 2
Nearly identical copy of CartoSat–2C.

CartoSat–2E ISRO 23.06.2017 ✔ 0,65 2
Fifth and last satellite of the Indian Cartosat.

FORMOSAT–5 NSPO 24.08.2017 ✔ 1 2
First indigenously developed remote sensing satellite of the National Space Organization of the Republic of China (Taiwan).

Back
3.1. Which resolution is High Resolution? Before analysing the potential archaeological utility of the data presented in Table 3, some premises seem necessary.

First of all, it must be specified why some satellites have been included whereas others have not. The criterion of "High Resolution" is not absolute: what was high resolution twenty years ago may not be considered as such right now .Thus, keeping in mind the technology leaps that happened in the last twenty years, this list includes sensors that, back when they were launched, were rightfully considered the cutting edge (or at least close to it) of technology, and can still give acceptable results today (i.e. with standing monumental archaeological remains in dry lands). As an example, we may consider EROS–A and SPOT 5: their ground sampling resolution, respectively 1,5 and 2,5m at nadir, could hardly be considered "high", if their scheduled launch was set tomorrow; back in 2002, though, when LANDSAT–5 and –7 had a maximum resolution of 30m at nadir, and the only other available products were top-class DigitalGlobe and GeoEye satellites, they could have served as a valid alternative (see the following chart).

Similarly, satellites launched in recent years that fail to give competitive results (approximately considered below the threshold of 1m ground resolution) have not been included. This is mainly due to a high possibility of a disadvantageous quality/price ratio: in fact, for the same price of new data commissioned to these satellites, better products may be found from other companies; if high level of detail is not strictly needed, old archive images can be bought, with no need of expensive new imagery. In this case we may consider Flock constellation, from Planet Labs inc. (who recently acquired TerraBella from Google). From January 9th, 2014 to July 14th, 2017 there have been 16 separate launches (Flock-1, -1b, -1c, -1d, -1d', -1e, -1f, -2, -2b, -2c, -2d, -2e, -2e', -2k, -2p, -3p), and in each launch a whole constellation of "doves" (nanosatellites, each measuring 30 × 10 × 10 cm) has been sent in orbit, for a grand total of 311 nanosatellites in three years and a half [4]. Indubitably, this is a remarkable result: a high number of small low-cost satellites that can monitor the whole Earth surface on a daily basis. But optical sensors fitted on the "doves" grant a maximum GSD of 3–5m: it is a level of detail that could be used for many purposes, but a little too low for archaeology, in 2017. That is why they have not been included in the diagram (fig. 3).

4. Archaeological potential. As it can be seen, an increasing number of non-military HR satellites for Earth observation have been launched in the past twenty years, with twelve of them launched in 2016 alone. The growth pattern of the number of HRS launched/in orbit can be seen in the following chart, updated to September 2017 (fig. 4).

Of the 41 HR satellites still in orbit, 29 can record images with a spatial resolution better than 1m/px. Without any doubt, the DigitalGlobe constellation stands out among them, both in spatial and spectral resolution: it owns all five existing commercial satellites that can take PAN pictures with a higher resolution than 0,5m/px (the four WorldViews and GeoEye–1), and both satellites that reach a ground sampling distance (GSD) <2m in MS imaging (WorldView–3 and –4. WorldView–3's MS sensor is also the best MS scanner in terms of band number (it can work in 8 visible/near infrared –V/NIR – bands, 8 SWIR, and 12 bands for climate control known as "CAVIS"). After DigitalGlobe come the European (Pléiades), Chinese, Korean, Indian and Israeli satellites, all of them capable of taking images with a resolution between 0,5 and 0,6m/px. But how many of these satellites can be effectively useful for archaeologists, and how much would it cost us to buy HR images from them? To answer this question, contacts have been made with different HRS owners, but only five of them replied to my inquiries (Table 3).

Table 3. HRS owners and relative imagery costs

Name Satellite(s) owned Price*

DigitalGlobe
WorldView (1 to 4)
GeoEye–1 Archive images can be browsed on the DigitalGlobe Imagefinder. Prices start from 17,5 US$/km² (+19% tax), with a minimum order of 25km² (GSD 50cm), with a minimum total of ≈ 500 US$.
New collections need a minimum order of 100km², and prices vary according to resolution, selected spectral bands, cloud cover, and product rendering and licences.

National Space Organisation (NSPO)
FORMOSAT–2, 5 A pricelist can be found here.
Prices around 2000 € per archive image.

ImageSat International (ISI)
EROS A (1,5m nadir) Archive: 0,5 US$/km², min. 200km²

EROS B (0,7m nadir) Archive (>60 days): 6 US$/km², min. 25km²
<60 days/New: 16 US$/km², min. 25/50km²

Centre National d'Études Spatiales (CNES)
SPOT (6, 7) (1,5m nadir) Archive: 2,9/3,8 €/km² (PAN/MS), min. 100km²
New: varies according to the number of observations, cloud cover and availability of the area of interest; for a single shot, 9,20 €/km², min. 500km²

Pléiades (1, 2) (0,5m nadir) Archive: 10 €/km², min. 25km²
New: see above; for a single shot, 56 €/km², min. 100km²

Canadian Space Agency (CSA)
RadarSat–2 Spotlight mode (single look complex, 1m): 6000 CAN$.
Ultra–fine inSAR (3m): 2040 CAN$/scene (20 x 20km), min. 5 scenes in 6 months.

Back

As it can be seen, pricelists for new imagery can often result prohibitive to many archaeological pockets. However, as the quality of the pictures rises, archive images, in some cases barely older than two months, become available at more affordable prices. For instance, a single 25 km² archive Pléiades shot, with a very good 0,5 m resolution, could cost no more than 16000 RUB / 250 €: not a cheap solution, but a lot less than the hundreds of thousands of roubles needed for the average new tasking imagery.

4.1. Alternative methods to get access to HRSI. However, apart from direct supply from the owner companies, there are several other ways to get HRSI. For example, there are third-party supplying companies. These companies often offer a high variety of datasets to choose from, and they sometimes remove the minimum order in km²; this results in more affordable costs, especially in small scale research areas. The most known third party supplying companies are Apollo Mapping, Harris Geospatial Solutions, Geocento, and Terra Server.

Finally, there are some methods for getting free access to High Resolution Satellite Imagery (HRSI). The first is the European Space Agency: ESA can grant free access to European Archive imagery datasets, after a proposal evaluation, as well as a possible access to the Third Party Mission Program, which allows access to archive and new tasking HRSI from multiple datasets, after the submission and evaluation of a highly detailed Research Project. The second is the previously mentioned Earth Explorer portal, from the United States Geological Survey: it grants free access to multiple medium resolution datasets such as LANDSAT and SENTINEL, and free or low budget access to declassified KeyHole imagery (Corona, Argos, Lanyard, Gambit, and Hexagon). The third method is represented by the most useful tool for low budget researches: Google Earth. Google Earth and similar softwares do not allow full access to the raw data, but to a composition of various satellite images, already processed and ready for free fruition. Nonetheless, it grants free exploitation of Pan-sharpened imagery sometimes as good as 0,4 m all around the world, as well as access to older satellite imagery (on the Desktop version): this is very useful for archaeological practice, since different periods of the year show different crop-markings. Google Earth Pro adds to the "classic" version of the product the possibility of measuring perimeters and extensions, as well as to get a very high level of image printing.

On version 9.0, released in April 2017, you can see only the most recent image taken for every single place, but, in addition, you get free access to 5m DEM on the majority of the Earth surface. Version 9.0 is not available to this day as a PC software, but can only be used through Chrome web browser and as an application for Android devices.

5. Conclusions. It has been shown how HRSI level is almost comparable to aerial photography: what twenty years ago looked like science fiction is increasingly becoming a matter of time. In fact, the number of orbiting satellites capable of giving potentially useful results for archaeological practice is progressively increasing, and the quality of optical payloads with it.

Unfortunately, High Resolution comes with high prices. But better sensors also mean better archives, and archive imagery is often significantly less expensive than new data. Among archives, declassified imagery still has an undeniable role, since it can often show at a very detailed level, as good as 0,5m/px, pre-industrial landscapes. Also, third party supplying companies own a broad selection of archive satellite imagery that is significantly less expensive than new data.

Finally, there are several ways to get HRSI at lower costs, and sometimes even for free. Among them, Google Earth (Pro) can still be considered the best way to get free access to HRSI.

References

Adams R.M., 1981. Heartland of cities: Surveys of ancient settlement and land use on the central floodplain of the Euphrates. Chicago: The University of Chicago Press. 362 p. Back

Alizadeh K., Ur J.A., 2007. Formation and destruction of pastoral and irrigation landscapes on the Mughan Steppe, north-western Iran // Antiquity. Vol. 81, iss. 331. P. 148–160. Back

Beck A., Philip G., Abdulkarim M., Donoghue D.N.M., 2007. Evaluation of Corona and Ikonos high resolution satellite imagery for archaeological prospection in western Syria // Antiquity. Vol. 81, iss. 331. P. 161–175. Back

Bitelli G., Girelli V.A, 2009. Metrical use of declassified satellite imagery for an area of archaeological interest in Turkey // Journal of Cultural Heritage. Vol. 10S. P. 35–40. Back

Casana J., 2004. The archaeological landscape of Late Roman Antioch // Culture and society in later Roman Antioch: Papers from a colloquium. London, 15th December 2001 / Ed. by I. Sandwell, J. Huskinson. Oxford: Oxbow Books. P. 102–125. Back

Casana J., 2007. Structural transformations in settlement systems of the Northern Levant // American Journal of Archaeology. Vol. 111, No. 2. P. 195–221. Back

Casana J., Cothren J., Kalayci T., 2012. Swords into Ploughshares: Archaeological Applications of CORONA Satellite Imagery in the Near East [Electronic resource] // Internet Archaeology. Vol. 32. URL: http://intarch.ac.uk/journal/issue32/2/toc.html. Back

Casana J., Cothren J., 2008. Stereo analysis, DEM extraction and orthorectification of CORONA satellite imagery: archaeological applications from the Near East // Antiquity. Vol. 82, iss. 312. P. 732–749. Back

Castrianni L., Di Giacomo G., Ditaranto I., Scardozzi G., 2010. High resolution satellite ortho-images for archaeological research: Different methods and experiences in the Near and Middle East // Advances in Geosciences. Vol. 24. P. 97–110. Back

Cerasetti B., Mauri M., 2002. The Murghab Delta palaeochannel reconstruction on the basis of remote sensing from space [Electronic resource] // Space Applications for Heritage Conservation. Proceedings. Strasbourg, 5–8 November 2002. CD-ROM. Back

Challis K., 2007. Archaeology’s cold war windfall. The Corona programme and lost landscapes of the Near East // Journal of the British Interplanetary Society. Vol. 60. P. 21–27. Back

Challis K., Priestnall G., Gardner A., Henderson J., O'Hara S., 2004. Corona remotely sensed imagery in dryland archaeology: The Islamic city of al-Raqqa, Syria // Journal of Field Archaeology. Vol. 29, no. 1/2. P. 139–153. Back

Comfort A., 1997. Satellite remote sensing and archaeological survey on the Euphrates // Archaeological Computing Newsletter. Vol. 48. P. 1–8. Back

Comfort A., Abadie-Reynal C., Ergeç R., 2000. Crossing the Euphrates in Antiquity: Zeugma Seen from Space // Anatolian Studies. Vol. 50. P. 99–126. Back

Commercial Satellite Imagery, 2002. A tactic in nuclear weapon deterrence / Ed. by B. Jasani, G. Stein. Berlin; Heidelberg: Springer–Verlag. 299 p. Back

Conesa F.C., Madella M., Galiatsatos N., Balbo A., Rajesh S.V., Ajithprasad P., 2015. CORONA Photographs in Monsoonal Semi-Arid Environments: Addressing Archaeological Surveys and Historic Landscape Dynamics over North Gujarat, India // Archaeological Prospection. Vol. 22, No. 2. P. 75–90. Back

Di Giacomo G., Scardozzi G., 2012. Multitemporal High-Resolution Satellite Images for the Study and Monitoring of an Ancient Mesopotamian City and its Surrounding Landscape: The Case of Ur // International Journal of Geophysics. Vol. 2012. P. 1–14. Back

Donoghue D.N.M., Galiatsatos N., Philip G., Beck A.R., 2002. Satellite Imagery for archaeological applications: a case study from the Orontes Valley, Syria // Aerial Archaeology. Developing Future Practice. NATO Advanced Research Workshop on Aerial Archaeology – Developing Future Practice (2000: Leszno, Poland) / Ed. by R. Bewley, W. Raczkowski. Amsterdam; Washington: IOS Press. P. 211–223. Back

Edwards D.N., 2010. Exploring rural landscapes in Sudanese Nubia // Landscapes through the lens. Aerial photographs and historic environment / Ed. by D. Cowley, R.A. Standring, M.J. Abicht. Oxford: Oxbow Books. P. 167–178. Back

Faustmann A., Palmer R., 2005. Wings over Armenia: Use of a paramotor for archaeological aerial survey // Antiquity. Vol. 79, iss. 304. P. 402–410. Back

Fowler M.J.F., 1996. High-resolution satellite imagery in archaeological application: a Russian satellite photograph of the Stonehenge region // Antiquity. Vol. 70, iss. 269. P. 667–671. Back

Fowler M.J.F., 1997. A Cold War spy satellite image of Bury Hill near Andover, Hampshire // Hampshire Field Club and Archaeological Society Newsletter. Vol. 27. P. 5–7. Back

Fowler M.J.F., 2001. A high-resolution satellite image of archaeological features to the south of Stonehenge // International Journal of Remote Sensing. Vol. 22, iss. 7. P. 1167–1171. Back

Fowler M.J.F., 2004a. Archaeology through the keyhole: the serendipity effect of aerial reconnaissance revisited // Interdisciplinary science reviews. Vol. 29, No. 2. P. 118–134. Back

Fowler M.J.F., 2004b. Declassified CORONA KH–4B satellite photography of remains from Rome’s desert frontier // International Journal of Remote Sensing. Vol. 25, iss. 18. P. 3549–3554. Back

Fowler M.J.F., 2008. The application of declassified KH–7 Gambit satellite photographs to studies of Cold War material culture: A case study from the former Soviet Union // Antiquity. Vol. 82, iss. 317. P. 714–731. Back

Fowler M.J.F., 2010. Satellite Imagery and Archaeology // Landscapes through the lens. Aerial photographs and historic environment / Ed. by D. Cowley, R.A. Standring, M.J. Abicht. Oxford: Oxbow Books. P. 99–110. Back

Fowler M.J.F., 2011. Modelling the acquisition times of Corona satellite photographs: Accuracy and application // International Journal of Remote Sensing. Vol. 32, iss. 23. P. 8865–8879. Back

Fowler M.J.F., 2013. Declassified Intelligence Satellite Photographs // Archaeology from Historical Aerial and Satellite Archives / Ed. by W.S. Hanson, I.A. Oltean. New York: Springer–Verlag. P. 47–68. Back

Fowler M.J.F., 2016a. A KH–9 satellite photograph of Soviet–era landscape graffiti // AARGnews: The newsletter of the Aerial Archaeology Research Group.No. 52 (March 2016). P. 87–89. Back

Fowler M.J.F., 2016b. The archaeological potential of declassified HEXAGON KH–9 panoramic camera satellite photographs // AARGnews: The newsletter of the Aerial Archaeology Research Group. No. 53 (September 2016). P. 30–36. Back

Fowler M.J.F., Fowler Y.M., 2005. Detection of archaeological crop marks on declassified Corona KH–4B intelligence satellite photography of southern England // Archaeological Prospection. Vol. 12, iss. 4. P. 257–264. Back

Gheyle W.R., Trommelmans R., Bourgeois J., Goossens R., Bourgeois I., De Wulf A., Willems T., 2004. Evaluating Corona: a case study in the Altai Republic (South Siberia) // Antiquity. Vol. 78, iss. 300. P. 391–403. Back

Goossens R., De Wulf A., Bourgeois J., Gheyle W.R., Willems T., 2006. Satellite imagery and archaeology: the example of CORONA in the Altai mountains // Journal of Archaeological Science. Vol. 33, iss. 6. P. 745–755. Back

Henderson J., Challis K., Gardner A., O’Hara S., Priestnall G., 2002. The Raqqa ancient industry project // Antiquity. Vol. 76, iss. 291. P. 33–34. Back

Hritz C., 2013. A malarial-ridden swamp: using Google Earth Pro and Corona to access the southern Balikh valley, Syria // Journal of Archaeological Science. Vol. 40, iss. 4. P. 1975–1987. Back

Kennedy D.L., 1998a. Declassified satellite photographs and archaeology in the Middle East: case studies from Turkey // Antiquity. Vol. 72, iss. 277. P. 553–561. Back

Kennedy D.L., 1998b. The twin towns of Zeugma on the Euphrates: rescue work and historical studies. Portsmouth, RI: JRA. 258 p. (Journal of Roman Archaeology. Supplementary series; No. 27). Back

Kouchoukos N., 2001. Satellite images and Near Eastern landscapes // Near Eastern Archaeology. Vol. 64, no 1/2. P. 80–91. Back

Min L., 2013. Archaeological Landscapes of China and the Application of Corona Images // Mapping archaeological landscapes from space / Ed. by D.C. Comer, M.J. Harrower. New York: Springer–Verlag. P. 45–54. Back

Moshier S.O., El-Kalani A., 2008. Late bronze age paleogeography along the ancient ways of Horus in Northwest Sinai, Egypt // Geoarchaeology. Vol. 23, iss. 4. P. 450–473. Back

Northedge A., 2005. Remarks on Samarra and the archaeology of large cities // Antiquity. Vol. 79, iss. 303. P. 119–129. Back

Oltean I.A., 2002. The use of satellite imagery for the transcription of oblique aerial photographs // Aerial Archaeology. Developing Future Practice. NATO Advanced Research Workshop on Aerial Archaeology – Developing Future Practice (2000: Leszno, Poland) / Ed. by R. Bewley, W. Raczkowski. Amsterdam; Washington: IOS Press. P. 224–232. Back

Palmer R., 2002. A poor man’s use of CORONA images for archaeological survey in Armenia [Electronic resource] // Space Applications for Heritage Conservation. Proceedings. Strasbourg, 5–8 November 2002. CD-ROM. Back

Palmer R., 2013. Uses of Declassified Corona Photographs for Archaeological Survey in Armenia // Archaeology from Historical Aerial and Satellite Archives / Ed. by W.S. Hanson, I.A. Oltean. New York: Springer–Verlag. P. 279–290. Back

Philip G., Donoghue D.N.M., Beck A.R., Galiatsatos N., 2002a. CORONA satellite photography: an archaeological application from the Middle East // Antiquity. Vol. 76, iss. 291. P. 109–118. Back

Philip G., Donoghue D.N.M., Beck A.R., 2002b. The contribution of satellite imagery to archaeological survey: an example from Western Syria [Electronic resource] // Space Applications for Heritage Conservation. Proceedings. Strasbourg, 5–8 November 2002. CD-ROM. Back

Quann J., Bevan B., 1977. The pyramids from 900 kilometres // MASCA Newsletter. Vol. 13. P. 12–14. Back

Reichhardt T., 2006. First Photo from Space [Electronic resource] // Air & Space Magazine. 24 Oct. 2006. URL: http://www.airspacemag.com/space/the-first-photo-from-space-13721411/?no-ist Back

Richards T.S., 1989. Evidence of rainwater concentrating structures in northern Egypt as seen on Landsat MSS imagery // International Journal of Remote Sensing. Vol. 10, iss. 6. P. 1135–1140. Back

Scardozzi G., 2010. The contribution of historical aerial and satellite photos to archaeological and geo-archaeological research: Case studies in Italy and Turkey // Advances in Geosciences. Vol. 24. P. 111–123. Back

Tappan G.G., Hadj A., Wood E.C., Lietzow R.W., 2000. Use of Argon, Corona, and Landsat Imagery to Assess 30 Years of Land Resource Changes in West-Central Senegal // Photogrammetric Engineering & Remote Sensing. Vol. 66, No. 6. P. 727–735. Back

Thompson H.O., 1967. A new development in archaeological air photography // Antiquity. Vol. 41, iss. 163. P. 225–227. Back

Trelogan J., Crawford M., Carter J., 2002. Monitoring the ancient countryside: remote sensing and GIS at the Chora of Chersonesos (Crimea, Ukraine) [Electronic resource] // Space Applications for Heritage Conservation. Proceedings. Strasbourg, 5–8 November 2002. CD-ROM. Back

Ur J.A., 2002. Settlement and landscape in Northern Mesopotamia: the Tell Hamoukar survey 2000–2001 // Akkadia. Vol. 123. P. 57–88. Back

Ur J.A., 2003. CORONA satellite photography and ancient road networks: a Northern Mesopotamian case study // Antiquity. Vol. 77, iss. 295. P. 102–115. Back

Ur J.A., 2005. Sennacherib’s Northern Assyrian canals: New insights from satellite imagery and aerial photography // Iraq. Vol. 67, iss. 1. P. 317–345. Back

Ur J.A., 2010. Urbanism and cultural landscapes in Northeastern Syria: The Tell Hamoukar Survey, 1999–2001. Chicago: University of Chicago Oriental Institute. 384 p. Back

Walstra J., Verkindren P., Heyvaert M.A., 2010. Reconstructing landscape evolution in the Lower Khuzestan plain (SW Iran): Integrating imagery, historical and sedimentary archives // Landscapes through the lens. Aerial photographs and historic environment / Ed. by D. Cowley, R.A. Standring, M.J. Abicht. Oxford: Oxbow Books. P. 111–128. Back

Wilkinson T.J., 2003. Archaeological Landscapes of the Near East. Tucson: The University of Arizona Press. 260 p. Back

Wilkinson T.J., French C.A.I., Matthews W., Oates J., 2001. Geoarchaeology, landscape and the region // Excavations at Tell Brak. Vol. 2: Nagar in the third millennium BC / Ed. by D. Oates, J. Oates, H. McDonald. Cambridge: McDonald Institute for Archaeological Research. P. 1–14. Back

Wilkinson T.J., Wilkinson E.B., Ur J.A., Altaweel M., 2005. Landscape and settlement in the Neo-Assyrian Empire // Bulletin of the American Schools of Oriental Research. Vol. 340. P. 23–56. Back

Wilkinson T.J., Wilkinson A.R., Philip G., 2006. Satellite imagery as a resource in the prospection for archaeological sites in central Syria // Geoarchaeology. Vol. 21, iss. 7. P. 735–750. Back

Notes

[1] I would like to deeply thank Dr. D. S. Korobov, who gave me opportunity to take part to the III International Conference “Archaeology and Geoinformatics”. Back

[2] A more detailed analysis of this first period of archaeological exploitation of satellites can be found in [Fowler, 2010]. Back

[3] G. Scardozzi analyses both Italian and Turkish data, and his paper has been counted in the six papers about Turkey [Scardozzi, 2010]. Back

[4] See here, for a general overview. For a more specific description of all launches, see single blog entries on the Planet Labs blog. Back

High Resolution Satellite Imagery and archaeology: a theoretical approach [1]

D. Danelli

Table 1.
Archaeological projects based on declassified satellite imagery (1998–2016)

Table 2.
List of commercial High Resolution satellites (HRS) updated to August 2017

Table 3. HRS owners and relative imagery costs

References

Notes

Place	Mission	Archaeological Use	Bibliography
Turkey	1103, 1104	First ever overhead photograph of the Roman city of Samosata on the Euphrates. Mapping, interpretation and illustration of key places, features, and geomorphology in the region of the Roman city of Zeugma prior to flooding of the area following the construction of the Birecik dam.	[Kennedy, 1998a; 1998b]
England	1104	Investigation of the archaeological potential of KH–4B imagery in Hampshire.	[Fowler, 1997]
Senegal	Corona Argon	Use of Corona and Argon imagery to assess environmental changes in Senegal between 1960s and 2000.	[Tappan et al., 2000]
Syria	Not specified	Aided geomorphological study of the complex landscape hinterland of the Bronze Age site of Tell Brak.	[Wilkinson et al., 2001]
Syria	1102, 1108, 1117	Identification of possible archaeological sites in the hinterland of Tell Hamoukar. The photographs were found to be extremely powerful for site identification since they preserve a landscape that is over thirty years old, prior to the expansion of towns, the intensification and mechanisation of agriculture and the introduction of diesel pumps and irrigated cotton fields.	[Ur, 2002]
Syria	1102, 1105, 1108	Identification and mapping of ancient road systems dating from the early Bronze Age in the Upper Khabur basin in north eastern Syria, where conventional aerial photography is limited.	[Ur, 2003]
Syria	1108, 1110, 1111	Aided identification of non-Tell settlement remains and distribution and organisation of field systems in Homs region.	[Philip et al., 2002a; 2002b; Donoghue et al., 2002]
Syria	1038	Provided broader landscape survey as part of an interdisciplinary investigation of an eighth to twelfth century Islamic industrial complex associated with the city of al-Raqqa in northern Syria. Predating both the modern urban expansion of Raqqa and development of extensive irrigation systems in the hinterlands, Corona photographs have proved invaluable for mapping the now largely destroyed cultural landscape.	[Henderson et al., 2002]
Iran	Not specified	Used together with Spot panchromatic imagery to illustrate specified landscape change on the Susiana plain in southwestern Iran between the late 1960s and early 1990s. During this period, much of the plain was levelled and an extensive network of irrigation channels constructed, which now mask traces of earlier irrigation systems of Sassanian and Islamic date that can be seen as dark lines on Corona photographs	[Kouchoukos, 2001]
Jordan	1115	Used to compare the appearance of features associated with Rome’s desert frontier visible on Corona images with conventional aerial photographs of the sites.	[Fowler, 2004b]
Ukraine	Not specified	Part of a multitemporal, multiresolution database of satellite imagery of Chersonesos, Crimea, an area that is one of the best preserved examples of an ancient Greek agricultural landscape. The site’s contemporary strategic position limits physical access and thus remote sensing offers an effective and reliable means of mapping the ancient landscape as well as monitoring urban encroachment over the past thirty years.	[Trelogan et al., 2002]
Turkmenistan	Not specified	Used as a dataset within a GIS to investigate settlement distribution relating to evolution of the ancient irrigation system of the river fan of the Murghab delta before the modern building of the Kara Kum canal, completed in 1988.	[Cerasetti, Mauri, 2002]
Armenia	1115	Used together with unsophisticated computer hardware and software to identify features including cultivation terraces, stone clearance cairns, enclosures, and sites of shepherds’ camps.	[Palmer, 2002]
Romania	1022	Used in place of large scale background maps for rectification and transcription of conventional oblique aerial photographs covering the Roman site of Micia.	[Oltean, 2002]
Near East	Not specified	Summary monograph on the archaeological landscapes of the Ancient Near East, which extensively exploits the Corona images.	[Wilkinson, 2003]
Syria	Not specified	KH–4B photographs were used to investigate the topography and cultural landscape of the early Islamic city of al-Raqqa in northern Syria.	[Challis et al., 2004; Challis, 2007]
Russian Federation	1042	KH–4A photographs were used as part of an archaeological survey of Bronze Age, Iron Age and ancient Turks monuments in the Altai Mountains of South Siberia	[Gheyle et al., 2004; Goossens et al., 2006]
Armenia	1115	KH–4B photographs were used together with digital photographs taken at low altitude from a paramotor were used to investigate a prehistoric, Medieval, and Soviet-era landscape in Armenia.	[Faustmann, Palmer, 2005]
Iraq	1102, 1108, 1117, 4017, 4031	KH–4B and KH–7 photographs were used to describe canals and other archaeological features related to water management and settlement in the landscape in the vicinity of the Assyrian city of Nineveh.	[Wilkinson et al., 2005; Ur, 2005]
England	1104	Ring ditches and a possible oval enclosure were detected as cropmarks on a KH–4B photograph acquired in August 1968.	[Fowler, Fowler, 2005]
Iraq	Not specified	Corona satellite photographs were used to supplement aerial photography to produce a photogrammetric plan of the ancient city of Samarra on the east bank of the Tigris.	[Northedge, 2005]
Turkey	1112	Used KH–4B photographs together with archaeological survey data to investigate aspects the settlement history of the Amuq Valley in the northern Levant including the archaeological landscape of Late Roman Antioch.	[Casana, 2004; 2007]
Iran	1103, 1110	Mapped using KH–4B photographs pastoral and irrigation landscapes in north-western Iran that have been virtually obliterated since the photographs were acquired.	[Alizadeh, Ur, 2007]
Iran	1052	KH–4A photographs were used to map the natural landscape of the Bushehr peninsula of Iran and to highlight changes to the landscape caused by modern water extraction and agriculture and to map individual monuments within the landscape.	[Challis, 2007]
Syria	1108, 1110	Archaeological sites captured on KH–4B were accurately located using a recent Ikonos satellite image of the Homs Region of western Syria and provides advice in the design and costing of surveys using satellite imagery.	[Wilkinson et al., 2006; Beck et al., 2007]
Egypt	1049	KH–4A photographs that revealed surface features that have been obscured by recent agricultural developments, together with field investigations of Holocene sedimentary deposits, were used to create a palaeogeographic map that places Late Bronze Age archaeological sites in Northwest Sinai in their environmental context.	[Moshier, El-Kalani, 2008]
Russian Federation	4028	A KH–7 photograph was used to describe the material culture of the Cold War installations around Moscow that were associated with the former Soviet Union’s first generation of surface to air missile (SAM) system.	[Fowler, 2008]
Turkey, Syria	1105	The stereo capabilities of KH–4B photographs were utilised to produce high-resolution digital elevation data and maps of archaeological landscapes in the Levant.	[Casana, Cothren, 2008]
Turkey	Not specified	Orthorectified KH–4A imagery was combined with LANDSAT Thematic Mapper, QuickBird and ASTER imagery to investigate the archaeological site at Tilmen Höyük (south-eastern Turkey).	[Bitelli, Girelli, 2009]
Syria	1046, 1105	The Tell Tuqan Survey Project used multi-temporal KH–4A and KH–4 photographs together with QuickBird imagery to study the tell and the surrounding area to inform the reconstruction of the ancient topography of the region.	[Castrianni et al., 2010]
Italy, Turkey	9022, 1043, 1049, 1103, 1107, 1109, 1111, 4036, 1210, 1216.	KH–3, KH–4A, KH–4B, KH–7 and KH–9 photographs were used to illustrate the importance of historical aerial and satellite photographs support archaeological and geo-archaeological research in Italy and Turkey.	[Scardozzi, 2010]
Iran	1035, 1045	The evolution of the Lower Khuzestan plain in SW Iran was investigated using KH–4A photographs, LANDSAT imagery together with archaeological, geological, and historical datasets.	[Walstra et al., 2010]
Sudan	Not specified	KH–4B and QuickBird imagery to support archaeological research and Cultural Resource Management in the Sudanese Middle Nile.	[Edwards, 2010]
Latvia	4031, 1110, 1210	KH–7, KH–4B and KH–9 photographs were used to describe the development of the Hen House ballistic missile early warning radars at Skrunda in Latvia between 1966 and 1975.	[Fowler, 2010]
Syria	1021, 1102, 1105 1108, 1117	Over 6.000 km of premodern trackways in the vicinity of Tell Hamoukar, one of the largest Bronze Age sites in northern Mesopotamia, were identified and mapped from KH–4A and KH–4B photographs.	[Ur, 2010]
Israel	1111, 1115	A comparison of two KH–4B photographs of the Roman siege-works at the fortress of Masada shows the impact of the solar lighting conditions at different times of the day on the appearance of the landscape and upstanding archaeological feature.	[Fowler, 2011]
Iraq	KH–4B e KH–7	Used KH–7, KH–4B, QuickBird–2, Ikonos–2 and WorldView–1 to study the ancient city of Ur, inaccessible since 2003.	[Di Giacomo, Scardozzi, 2012]
Near East	Not specified.	Declassified satellite imagery was used to study environmental and climatic changes, and to map new archaeological discoveries obtained through satellites in vast portions of the Middle East.	[Casana et al., 2012]
Syria	Not specified	Use of declassified satellite imagery to study places where political choices and/or deep vegetation prevent other ways to acquire aerial data.	[Hritz, 2013]
China	1101, 1115	Study about potentiality of Corona imagery on Chinese soil, focusing in particular on the ancient capital city of Qufu and on the post-war agricultural development.	[Min, 2013]
Armenia	1115	Use of KH–4B imagery taken in Armenia on the 20th September 1971 to research potential archaeological evidence.	[Palmer, 2013]
India	1117	Use of KH–4B imagery to survey monsoon floodplains in northern Gujarat and to analyse the evolution of agricultural and industrial landscape.	[Conesa et al., 2015]
Russian Federation	Not specified	On a KH–9 picture of the Kapustin Yar missile base East of Volgograd it was possible to see a landscape graffiti depicting the name ЛЕНИH .	[Fowler, 2016a]
USA	1217-4	Analysis of landscape and archaeological features of a picture taken from the primary camera of a KH–9 Hexagon satellite representing the city of St. Augustine in Florida, where the Spanish fort Castillo de San Marcus was built between 1672 and 1695.	[Fowler, 2016b]

Name	Producer	Launch date	In orbit	PAN (m)	MS (m)	Notes
EarlyBird–1	EarthWatch	24.12.1997	-	3	15	EarlyBird–1 was the first satellite ever to be built with the sole goal of achieving high resolution images for commercial purpose. Unfortunately, all contacts were lost 4 days after launch, and no image was taken.
Ikonos	EarthWatch	24.09.1999	-	0,81	3,2	Originally named Ikonos–2, it was re-named after the launch failure of Ikonos–1 on April 27th, 1999. In orbit until March 31st, 2015.
EROS A	ISI	05.12.2000	✔	1,5	/	First Israeli commercial satellite, produced by Israeli Aerospace Industry.
OrbView–4	GeoEye	21.09.2001	-	1	4	Launch failure. First satellite ever equipped with hyperspectral scanners. Ground resolution of OV–1 & –2 was more than 1km, and OV–3 was launched two years later than OV–4. DigitalGlobe bought GeoEye in January 2013.
QuickBird–2	DigitalGlobe	18.10.2001	-	0,61	2,44	EarthWatch inc. became DigitalGlobe inc. in September 2001. QB–1's launch was not successful. QB–2 stayed in orbit until January 24th, 2015.
SPOT–5	CNES	03.05.2002	-	2,5	10	Fifth Satellite Pour l'Observation de la Terre of the French Centre National d'Études Spatiales. His older brother, SPOT–4, in orbit from 1998 to 2013, had a 10m PAN and 20m MS ground resolution. SPOT–5 stopped working on March 31st, 2015.
OrbView–3	GeoEye	26.06.2003	-	1	4	No image was taken after April 23rd, 2007. On March 13th, 2011, the satellite was taken down.
FORMOSAT–2	NSPO	21.05.2004	-	2	8	First HRS from Republic of China (Taiwan)'s National Space Organization. Its scanners could achieve 2m PAN images and 8m 4-bands multispectral images. After twelve years, it was decommissioned on 1st Aug 2016.
CartoSat–1 (IRS–P5)	ISRO	05.05.2005	✔	2,5	/	First HRS produced by ISRO, India Space Research Organisation. It was equipped with a panchromatic camera only.
EROS B	ISI	25.04.2006	✔	0,6	/	Second HR satellite launched by Israeli Space Agency.
Resurs–DK–1	NTs OMZ	15.06.2006	-	1	2–3	First HR civil satellite of Russian space agency. It remained in orbit longer than expected, and communications finally terminated on February 17th, 2016.
CartoSat-2	ISRO	10.01.2007	✔	0,8	/	Second Indian HRS, first to have a better resolution than 1m²/px.
WorldView–1	DigitalGlobe	18.09.2007	✔	0,41	/	DigitalGlobe's WorldView, at the moment the constellation of satellites with the highest resolution, was the first to achieve a resolution better than 0,5m. However, until June 2014, only US Government could access the maximum resolution, and denied the commercial use of imagery with a resolution better than 0,5m.
RadarSat–2	CSA	14.12.2007	✔	3	/	First HR satellite of the Canadian Space Agency, and first synthetic-aperture radar (SAR) ever to get to the threshold of 3m ground resolution(RadarSat–1's ground resolution was 10x9 meters at its best).
CartoSat–2A	ISRO	28.04.2008	✔	0,8	/	Third Indian HRS.
GeoEye–1 (OrbView–5)	GeoEye	06.09.2008	✔	0,41	1,65	The US government prohibition to sell imagery with a resolution better than 0,5m was also in force for GeoEye, later acquired by DigitalGlobe itself. First MS imagery with ground resolution better than 2m.
WorldView–2	DigitalGlobe	08.10.2009	✔	0,41	1,85	First satellite of the WorldView constellation equipped with a MS scanner.
CartoSat–2B	ISRO	12.07.2010	✔	0,80	/	Fourth HRS of the Indian constellation.
Pléiades–1A	CNES	17.12.2011	✔	0,5	2,8	First of a two HRS constellation commissioned to the Centre Nationale d'Études Spatiales, SPOT satellites producers, by the ESA, European Space Agency.
KompSat–3	KARI	27.05.2012	✔	0,7	2,8	First HRS of the Korea Aerospace Research Institute, South Korean space agency.
SPOT–6	CNES	09.09.2012	✔	2	8	Sixth French Satellite Pour l'Observation de la Terre, ten years after the launch of SPOT–5.
Pléiades–1B	CNES	02.12.2012	✔	0,5	2,8	Second of the two Pléiades satellites.
GaoFen–1	CNSA	26.04.2013	✔	2	8	First HRS of the Chinese National Space Agency. Gao Fen in Chinese means high resolution.
Resurs–P1	Roscosmos	25.06.2013	✔	1	3–4	Second Russian HRS.
SkySat–1	Skybox Imaging (then TerraBella)	21.11.2013	✔	0,9	2	Skybox Imaging, founded in 2009, launched its first HRS back in 2013. Less than a year later (Aug 2014), the company was acquired by Google inc. for half a million USD, and re-named Terra Bella (Italian for Beautiful Earth), from Terra Bella ave., Mountain View, CA, where Google is based.
SPOT–7	CNES	30.06.2014	✔	2	8	Seventh French Satellite Pour l'Observation de la Terre .
SkySat–2	Skybox Imaging	08.07.2014	✔	0,9	2	Second Skybox/Terra Bella HRS.
WorldView–3	DigitalGlobe	13.08.2014	✔	0,31	1,24	The third WorldView satellite, along with the fourth, are equipped with the best optical scanners available on market. Until Feb. 2015 US government forbade the selling of the maximum resolution images. Its MS scanner can record images in 16 bands (six in visible light, two NIR from 770 to 1040nm, and eight shortwave infrared – SWIR – from 1195 to 2365nm) at a maximum resolution of 1,24m. On top of that, WorldView–3 is also equipped with a CAVIS scanner (Clouds, Aerosols, Vapers, Ice and Snow) that records images in 12 bands from 405 to 2245nm, with a ground resolution of 30m.
GaoFen–2	CNSA	19.08.2014	✔	0,8	3,2	Second Chinese HRS, first with a resolution better than 1m.
Resurs–P2	Roscosmos	26.12.2014	✔	1	3–4	Third Russian HRS .
KompSat–3A	KARI	25.03.2015	✔	0,55	2,2	Second Korean HRS.
GaoFen–8	CNSA	26.06.2015	✔	?	?	Probably the best HRS in the Chinese constellation, although technical data have not been released.
TripleSat	SSTL	10.07.2015	✔	0,8	3,2	HRS triplet with better than 1m resolution launched by Surrey Satellite Technology Ltd., associated with Surrey University.
GaoFen–9	CNSA	14.09.2015	✔	0,5	2	Chinese HRS with better known resolution.
Rersus–P3	Roscosmos	13.03.2016	✔	1	3–4	Fourth Russian HRS.
CartoSat–2C	ISRO	22.06.2016	✔	0,65	2	Best Indian HRS at the moment, with a resolution of 0,6m PAN and 2m MS.
SkySat–3	TerraBella	22.06.2016	✔	0,9	2	First SkySat launched under Google aegis, on the same missile that carried CartoSat–2C.
PerúSat–1	CONIDA	16.09.2016	✔	0,7	2	First satellite of the Comisión Nacional de Investigación y Desarrollo Aeroespacial, Peruvian Space Agency.
SkySat–4/–7	TerraBella	16.09.2016	✔	0,9	2	On the same missile that carried PerúSat–1 there were also 4 SkySat satellites of the Terra Bella/Google HR constellation. In the future, said constellation is to be composed of 21 elements. On 3rd February 2017, Planet Labs inc. stated on its website to have acquired TerraBella from Google.
WorldView–4	DigitalGlobe	26.11.2016	✔	0,31	1,24	The launch originally scheduled for June and then for September 2016, was re-scheduled due to the wildfire that burned the woods surrounding the base of Vanderberg, CA, where the launch was set. Along with WorldView–3, it is the satellite which takes the images with the best resolution on the market.
Göktürk–1	Turkish Aerospace Industries	05.12.2016	✔	0,8	/	The satellite, commissioned by the Turkish Ministry of Defence to the Italian company Telespazio, was the first Turksih HRS to break the barrier of 1m PAN resolution. His “older brother”, Göktürk–2, was launched on Dec 12th, 2012, and it can take images with a resolution of 2,5m PAN, 10m in four MS bands and 20m in an experimental SWIR band.
GaoJing–1/–2	CNSA	28.12.2016	✔	0,5	2	The two GaoJing minisatellites, first steps of a newer and wider Chinese constellation, were launched on Dec 28th, 2016. However, mistakes were made during the launch, and part of the missile payload crashed to the ground. After a couple of weeks of uncertainty about their fate, the two satellites became operative on Jan 12th, 2017.
CartoSat–2D	ISRO	15.02.2017	✔	0,65	2	Nearly identical copy of CartoSat–2C.
CartoSat–2E	ISRO	23.06.2017	✔	0,65	2	Fifth and last satellite of the Indian Cartosat.
FORMOSAT–5	NSPO	24.08.2017	✔	1	2	First indigenously developed remote sensing satellite of the National Space Organization of the Republic of China (Taiwan).

Name	Satellite(s) owned	Price*
DigitalGlobe	WorldView (1 to 4) GeoEye–1	Archive images can be browsed on the DigitalGlobe Imagefinder. Prices start from 17,5 US$/km² (+19% tax), with a minimum order of 25km² (GSD 50cm), with a minimum total of ≈ 500 US$. New collections need a minimum order of 100km², and prices vary according to resolution, selected spectral bands, cloud cover, and product rendering and licences.
National Space Organisation (NSPO)	FORMOSAT–2, 5	A pricelist can be found here. Prices around 2000 € per archive image.
ImageSat International (ISI)	EROS A (1,5m nadir)	Archive: 0,5 US$/km², min. 200km²
ImageSat International (ISI)	EROS B (0,7m nadir)	Archive (>60 days): 6 US$/km², min. 25km² <60 days/New: 16 US$/km², min. 25/50km²
Centre National d'Études Spatiales (CNES)	SPOT (6, 7) (1,5m nadir)	Archive: 2,9/3,8 €/km² (PAN/MS), min. 100km² New: varies according to the number of observations, cloud cover and availability of the area of interest; for a single shot, 9,20 €/km², min. 500km²
Centre National d'Études Spatiales (CNES)	Pléiades (1, 2) (0,5m nadir)	Archive: 10 €/km², min. 25km² New: see above; for a single shot, 56 €/km², min. 100km²
Canadian Space Agency (CSA)	RadarSat–2	Spotlight mode (single look complex, 1m): 6000 CAN$. Ultra–fine inSAR (3m): 2040 CAN$/scene (20 x 20km), min. 5 scenes in 6 months.

High Resolution Satellite Imagery and archaeology: a theoretical approach [1]

D. Danelli

Table 1. Archaeological projects based on declassified satellite imagery (1998–2016)

Table 2. List of commercial High Resolution satellites (HRS) updated to August 2017

Table 3. HRS owners and relative imagery costs

References

Notes

Table 1.
Archaeological projects based on declassified satellite imagery (1998–2016)

Table 2.
List of commercial High Resolution satellites (HRS) updated to August 2017