A Cross-Section of Results from the 2015 Lidar Campaign
A lot of the preliminary results and analyses of the 2015 archaeological lidar campaign will be coming online in the coming months. An initial peer-reviewed publication will appear in the coming weeks in the Journal of Archaeological Science, which outlines the most significant outcomes, and there will be a talk at the Royal Geographical Society in London on the 13th of June 2016 that will cover many of the same issues.
Tickets to the London event can be purchased here, and all proceeds will go to funding the work of Global Heritage Fund (GHF) at the Cambodian temple complex of Banteay Chhmar, which we also covered with lidar and where GHF supports a number of programs including an important Community Based Tourism initiative.
In the meantime here are some examples of the terrain models, stripped of vegetation, derived from the new dataset (click for larger detail):
The largest enclosure in pre-modern Southeast Asia is at the site of Preah Khan of Kompong Svay, where my colleague Mitch Hendrickson – who completed his doctorate at the same time as me at the University of Sydney, and is now on faculty at the University of Illinois at Chicago – has been working for about a decade in association with the Ministry of Culture and Fine Arts.
The site is largely obscured by open (and largely deciduous) woodland but the lidar deals very well with this kind of environment when we fly in ‘leaf-off’ conditions in March/April, as we did last year. A field season of excavation has already (successfully) addressed the extent to which surface lidar returns are reflected in subsurface remains.
The lidar company (PT McElhanney Indonesia) who did our acquisition for CALI did a completely separate acquisition in Preah Vihear Province under the administrative purview of the Ministry of Mines and Energy, which happily coincides with several areas that saw iron mining and smelting during the Angkor period (and perhaps before and after). They were kind enough to let us analyse the data for archaeological remains, and it turns out that ancient iron-working sites have an extremely clear signature in the lidar data, even though the data quality (density of points) is about a quarter of what we’d normally order for an archaeological project.
You can see in the images above a very distinctive pattern of iron smelting furnaces (the small, pronounced mounds) surrounding what appears to be a pond, but which we theorise may in fact be infilled iron mines that turned into ponds.
Contrary to an implication made in the Phnom Penh Post that lidar was somehow missing iron smelting sites, it turns out that we get beautiful images of smelting sites, which has already been confirmed on the ground in joint projects with the Ministry of Culture and Fine Arts in Preah Vihear Province this year (in fact, the site mentioned in the Post article has never actually been overflown with a lidar).
The 2015 campaign vastly expanded the lidar coverage of the remote and mountainous regions to the north of Angkor, not only to get a complete coverage of Phnom Kulen but also the extended mountain range to the west (a mountain range which incidentally seems to have no collective name, just individually-named peaks).
The so-called “lost city” of Mahendraparvata does not extend beyond the plateau of Phnom Kulen proper, it turns out, which is not to say that the rest of the coverage is uninteresting: we’ve found a huge number of new temples, ancient dams, ponds, temples, quarries, and other evidence of Angkor-era expansion into these ranges. Several of these features can be identified in the image above, if you know what you’re looking for….
I mentioned in an earlier post that the results at Sambor Prei Kuk are quite spectacular. You can compare the raw unprocessed data with the publication-quality images that we now have in hand. Thanks to many years of hard work on behalf of the joint Cambodian-Japanese team at Sambor Prei Kuk we already had some very nice plans of the layout of the central temples, which you can see clearly in this image. The lidar data adds a whole new dimension though, showing a quite complex system of moats, waterways and other features that had not been mapped in detail before. This is just the central few sq km of the Sambor Prei Kuk data; we actually acquired about 200 sq km over the site and its environs. The data turns out to be extremely valuable archaeologically, and should add a new dimension to our understanding of the development of this pre-Angkorian capital and how it probably persisted as an important center well into the Angkor era.
One of the great benefits of high-accuracy, high-resolution topographic data is the ability to very precisely model water flows not only in modern field systems but also in pre-modern ones (which very often remain in use in the present day). This helps us to model the evolution of hydrology in a given area, in order to understand how water management systems functioned (or sometimes failed to function!) in times past, as well as today. This lidar image is taken in the town of Oudong, just north of Phnom Penh, in which we can see very clearly the subtle gradations between elevated areas and inundated areas. This area was a capital of the Khmer during the so-called ‘Middle Period’ (~14th-19th centuries AD) and has now been pretty comprehensively covered by modern development – for example Highway 5 from Phnom Penh to Battambang, which runs horizontally across the center of the image. Nonetheless the processing algorithms we use are smart enough to filter out a lot of the modern development and reveal a system of pre-modern water management including reservoirs, dams and other features.
The site of Choeung Ek lies within the greater Phnom Penh area and has been the subject of a lot of media attention lately due to the threats posed by modern development to this very important archaeological site. There was an excellent article about it in the Cambodian Daily about it recently following on from another article in the Phnom Penh Post last year. What you can see above is conventional aerial imagery of the site alongside lidar imagery in which the vegetation and modern development has been mostly removed. You can see the circular moated site a lot more clearly in the lidar, and the destruction of the site is pretty striking. This site was not actually included in our original campaign planning, but we included it during the instrument calibration phase at the request of Phon Kaseka, our esteemed colleague from the Royal Academy of Cambodia.
You can see where the lidar has difficulty with waterlogged terrain and dense low-lying scrub in this kind of testing mode; if we’d flown with full density and/or with waveform the picture would’ve looked a lot smoother – even so, the lidar is a great improvement on conventional imagery….
Disclaimer: All views and opinions expressed on this page are those of the author and do not necessarily represent official positions of APSARA, the EFEO or the Ministry of Culture and Fine Arts.