Lidar Reveals Hidden Angkor Ruins

An aerial lidar survey provides archaeologists with new insight to help uncover the secrets of Angkor’s ancient civilization in Cambodia.

By Francisco Gonçalves

An extensive lidar and aerial photography acquisition project has provided insight into the former Khmer Empire, the major power in Southeast Asia between the 9th and 15th centuries. The ruins of its capitals lie in Angkor, within present-day Siem Reap Province, Cambodia. Believed to have been the largest pre-industrial urban complex in the world, Angkor may have supported a population of up to one million people at its peak. Its most famous temple, Angkor Wat, is depicted on the Cambodian flag. Angkor is one of the most important archaeological sites in Southeast Asia and was designated a UNESCO World Heritage Site in 1992.
 

Remote Sensing in Archaeology

Archaeologists have often relied on the use of technologies developed for other purposes to support their research. Beginning with aerial reconnaissance and photography in the 1920s, remote sensing has offered researchers an effective method of non-destructively surveying large areas. Aerial photography of Angkor was acquired throughout the 1920s and 1930s, and a combined aerial and land-based mapping project was initiated in the 1950s; it was left uncompleted due to funding limitations and political turmoil. After nearly two decades of unrest, archaeological research resumed in Angkor in 1992. A series of radar surveys was undertaken between 1994 and 2000, most significantly AIRSAR/TOPSAR data acquired by NASA in September 2000.

In April 2009, lidar was successfully employed by archaeologists in the ancient Maya city of Caracol, in modern Belize. Although lidar had been used previously at other historical sites—including the Stonehenge World Heritage Site—this was the first archaeological application over such a large area. The data confirmed and surpassed the results of over 25 years of field study, revealing a vast settlement, 200 sq km in area, with a heavily modified landscape.
 

Bringing Lidar to Southeast Asia

Upon learning of the success of lidar in Caracol, PT McElhanney Indonesia (PTMI), a division of McElhanney Consulting Services Ltd., was inspired to bring the technology to archaeological projects in Southeast Asia. Important historical regions such as Siem Reap and Ayutthaya are believed to have many ancient wonders yet to be discovered. It is estimated that, in Siem Reap alone, only a fraction of the ancient infrastructure has been unearthed thus far. The use of lidar would expand on previous surveys to reveal more detailed features of the landscape.

Radar imaging in Angkor had provided evidence of new settlements, canals, and reservoirs, but was less effective at detecting smaller features and low constructions. Its use was also very limited over the extensive forested areas around Angkor. While radar produced an elevation value for every five square meters of landscape, lidar would generate multiple data points within a single square meter, even penetrating the dense jungle canopy. Its accuracy far exceeds that of other remote-sensing technologies, providing a detailed picture of the topography and modifications made by ancient civilizations. Very subtle traces of previous development can still be detected on the surface of the landscape hundreds or even thousands of years later.

To advance the idea of using lidar within Southeast Asia, PTMI contacted professor Roland Fletcher of the University of Sydney’s archaeology department, the principal director and instigator of the Greater Angkor Project. We were able to arrange for a meeting and presentation in Siem Reap to discuss how lidar could assist archaeologists working in the region. Prior to this, we met with Dr. Damian Evans, director of the Robert Christie Research Centre and deputy director of the Greater Angkor Project, to evaluate the pros and cons of using lidar on tropical forest-covered terrain and to determine what archaeologists hoped to achieve.

Dr. Evans was aware of the work done in Caracol and did not expect the same results in Siem Reap due to their differing landscapes. Unlike Caracol, which lies beneath undisturbed forest, much of Siem Reap comprises secondary forests, agricultural land, and urban developments. However, it was expected that lidar would, at a minimum, delineate roadways, reservoirs, and similar features.

Following our meeting with Dr. Evans, we conducted a presentation at the Siem Reap Centre, the research base of the École Française d’Extrême Orient. Several archaeological officials attended, representing at least 15 separate research organizations from around the world. Although there was great interest in employing the technology, it soon became clear that the primary bottleneck would be in raising the necessary funds to allow such a project to proceed.

At PTMI’s suggestion, a consortium was formed among eight of these groups, who agreed to share the data as well as the cost of acquisition. The Khmer Archaeology Lidar Consortium (KALC) would be headed by Dr. Evans, with the following organizations participating:

  • Authority for the Protection and Management of Angkor and the Region of Siem Reap (APSARA) [Cambodia]
  • École Française d’Extrême Orient (EFEO), Siem Reap Centre [France]
  • University of Sydney, Robert Christie Research Centre (USYD) [Australia]
  • Société Concessionnaire des Aéroports (SCA) [France and Cambodia]
  • Hungarian Indochina Company (HUNINCO) [Hungary]
  • Archaeology & Development Foundation (ADF), Phnom Kulen Program [France]
  • Japan-APSARA Safeguarding Angkor (JASA) [Japan]
  • World Monuments Fund (WMF) [USA].
 
APSARA would act as government lead, the EFEO as administrative lead, and the University of Sydney as technical lead. A proposal was developed whereby PTMI would acquire lidar data with a minimum of two points per square meter and aerial photography with a ground pixel size of approximately 15 cm. The survey would cover an estimated 270 square kilometers over three major blocks: Angkor and Phnom Kulen in Siem Reap Province and Koh Ker in Preah Vihear Province.

Surveying the Ruins

Prior to mobilization of the lidar system and equipment, PTMI traveled to Siem Reap to conduct a field assessment. We spent two days visiting the various project sites to better understand the archaeologists’ requirements and what they hoped to gain from the data. This proved invaluable when developing flight plans and specifications.

Shipping the lidar system presented challenges, as it was difficult to find an agent that could handle the size of the pod. The archaeologists and our helicopter owner-operator assisted with obtaining shipping clearances and negotiating import exceptions, and we were able to arrange for shipment to Phnom Penh, Cambodia’s capital city. The pod travelled by air to Singapore, then by ground to Phnom Penh via Thailand. Once in Phnom Penh, the pod, lidar system, and camera system were mounted on the helicopter, and all systems were tested and calibrated.

The acquisition was completed over a seven-day period of intense heat, scheduled between foggy mornings, afternoon cloudy periods, and Siem Reap’s regular air traffic. Each evening, the data was checked for coverage and preliminary quality, with preview images generated and sent to the archaeologists. The overall results were even better than predicted, with billions of data points collected, complemented by approximately 5,000 digital aerial photographs. Prior to delivery to the KALC, the data points were processed and classified as either ground or non-ground, and the aerial photographs were color balanced and indexed.

The data was collected using a Leica ALS60 airborne laser scanner and a Leica RCD105 digital frame camera mounted on an AS350 B2 helicopter. The survey was undertaken at an airspeed of 85 knots, at a flying height of 800 meters above ground level. The ALS60 was set at a pulse rate of 120 kHz, with full waveform acquired. We used a swath width of 670 meters, achieving an average point density of 10-15 points per square meter before classification, with photo frames taken every 215 meters along the flight lines.
 

The Future of Lidar in Archaeology

According to Dr. Evans, the lidar survey has produced “a lifetime” of data—in only one week. More than 2 TB of data have now been delivered for further manipulation and analysis. Lidar’s ability to penetrate dense vegetation and to detect even subtle elevation differences has revealed a vast array of previously undocumented features such as roadways, canals, urban enclosures, and temples. This will aid researchers in developing a more detailed map of the site, one of the primary goals of the Greater Angkor Project.

Preliminary analysis points to important new discoveries that, according to archaeologists in Siem Reap, will revolutionize the history of the Khmer Empire. Of particular note is a new settlement discovered surrounding Beng Mealea, a known temple between Angkor and Koh Ker. The area is obscured by forest cover, and the presence of land mines has made ground survey both difficult and dangerous. Lidar has revealed the area as an intricate, planned settlement with a network of gridded roadways and ponds. It has been impeccably preserved, offering researchers a first-hand view of a 12th century Khmer city.

The findings in Siem Reap have provided substantial insight into the Khmer civilization as well as temple complexes worldwide. Many known temple sites may actually be the centers of low-density urban areas, currently enshrouded in vegetation or covered by modern settlement. Even in areas studied extensively, lidar can detect previously undiscovered features or provide a new level of detail.

In addition to its contributions to archaeological research, the use of lidar has significant implications for site and heritage management. Lidar data can identify areas requiring immediate protection, create baseline and archival documentation, and warn of illegal activities such as excavation, looting, and deforestation. Insight may also be gained into the scale and extents of historical settlements to more accurately assess zone boundaries, to ensure these important sites are protected and preserved well into the future.


Francisco Gonçalves is the president director of PT McElhanney Indonesia, a division of McElhanney Consulting Services Ltd., based in Vancouver, Canada. Gonçalves has an extensive background in geomatics and GIS, with over 30 years’ experience in mapping projects around the world and 20 years of operations in Southeast Asia.
 

 

» Back to our Aerial Mapping Fall '12 Issue

Website design and hosting provided by 270net Technologies in Frederick, Maryland.