In Search of Civilization: Using Modern Technology to Discover Ancient Treasures

John Urwin | john.urwin@yale.edu November 18, 2011

Archaeologists typically encounter dense understory while surveying. All images in this article are courtesy of Carlos Chiriboga.

Looking at satellite photographs of the Guatemalan wilderness in the basement of 51 Hillhouse Avenue, the home of the Yale Department of Archeology, I could not help but think of how trying to find ancient ruins within the sea of green is like trying to find a single needle in a six hundred thousand acre jungle. The task seemed daunting. Not only were the ruins hidden from above, but they were also held within a dangerous and remote wilderness. But where I saw a challenge, Anthropology graduate student Carlos Chiriboga sees opportunity.

As part of the Proyecto Regional Arqueólogico La Corona (PRALC), Chiriboga has ventured into the Petén Lowlands of the southern Maya lowlands twice for his dissertation research. After flying into Guatemala City in March, Chiriboga and his team of surveyors drive five hours in a specially equipped 4×4 double cab truck to their base camp of Buena Vista. From this camp, they continue to venture out into the jungle in search of the remains of a glorious civilization long lost to time.

Back in ancient times, these lowlands and alluvial basins were home to the once flourishing Maya civilization. On this same soil, the Maya people developed a society that grew to the sophistication and beauty that we know of today. By the Classic Period (c. 200-900 C.E.), this region had solidified itself as the heartland of the empire, boasting the storied urban centers of Tikal, Uaxactun, San Bartolo, Naranjo, Holmul, and La Sufricaya and a population of several million. Doubtlessly, the area was one of the most densely populated and cultured regions in the entire world.

Today, however, the land has become largely emptied of man’s presence. As Chiriboga’s team of surveyors and excavators journeys through the thick jungle underbrush, reminders of the power of nature surround them. The land where the mighty civilization once stood is now overgrown with trees, overrun by nature. After prosperity throughout the Classic Period, Maya civilization went into decline during the 8th and 9th centuries until its eventual collapse. As time went on, populated settlements were abandoned, projects for new monuments were halted, and the remnants were left to erode with time and become repossessed by the jungle’s thick vines.

What little human presence remains in these remote areas of the Northwest Petén tends to be linked to the international drug trade, a fact constantly on the mind of Chiriboga and his team. Despite numerous leads indicating promising ruins in the western corner of his survey region, he cannot explore those parts, as illegal settlers, loggers, and drug smugglers dominate these lands. In fact, it is possible to visualize their disruptive presence by tracking the proliferation of small airstrips slashed out of the jungle to support their contraband-laden aircrafts.

In addition to these issues, when his team does reach a new site, they usually find it to have been previously sacked and looted. Throughout the early 20th century, local residents, known as chicleros, would annually venture into the wilderness to harvest chicle, the main ingredient in chewing gum, and sell the crop upon their return. On these expeditions, they would pass by ruins and note their locations, only to return during the off-season to loot them. To this day, many archeological sites are actually “discovered” by asking former chicleros. In fact, Chiriboga’s main surveyor and workman, Don José Luís Morales, is a 75-year-old ex-chiclero. But despite this silver lining, these practices have left few intact ruins. Invaluable treasures have been stolen, and the once beautiful masonry lies ransacked, appearing, in Chiriboga’s words, “as if a bomb had gone off.”

Looting has damaged almost all of the sites encountered by Chiriboga’s team. Even small structures, like the one in this image, have been the subject of such practices.

Even so, the value of the treasures possessed by the Petén Lowlands is undeniable. According to the Guatemalan Ministry of Culture, researchers have already discovered 500 archeological sites, with another potential 4,500 thought to be yet undiscovered. Included in those already discovered are the famous archeological sites of Tikal, Holmul, and La Sufricaya.

Once an aggressive and expansionist state of pre-Columbian Maya civilization, Tikal, or what remains of it today, rests in the shadows of its monumental temple, which rises over 150 feet above its Great Plaza. As United Nations Educational, Scientific and Cultural Organization (UNESCO) describes it, “The ruined city reflects the cultural evolution of Mayan society from hunter-gathering to farming, with an elaborate religious, artistic, and scientific culture.” Holmul, for its part, holds to its credit continuous settlement for nearly two millennia, from 800 B.C.E. to around 900 C.E, offering a wide spectrum of historical development and demise. La Sufricaya also has left us with beautiful tablets and murals, along with structures including an acropolis, public buildings, a ball-court, and funerary temples.

Yet, for all that these jungles hold, they remain largely unexplored. While the region due east of Chiriboga’s has been the subject of fairly intensive archeological research, the northwest quadrant has been mostly ignored. Relying on modern mapping technology and his own site prediction model, Chiriboga aims to change this by completing a survey and mapping the location of the major Maya settlements in this region of nearly 1,000 square miles of thick, dangerous jungle.

Under usual circumstances, Chiriboga would proceed by conducting one of two types of “traditional” surveys: either a random sample survey or a full-coverage survey. After selecting a region of interest and dividing it into either quadrats (50 m x 50 m, 100 m x 100 m, or 500 m x 500 m, for instance) or transects (100 m x 1000 m, etc.), he would proceed to physically survey the area. In a random survey, he would choose a minimum number of quadrats or transects to survey (the exact number would be determined by following statistical formulas). He would then proceed to the actual survey, in which he and his group of surveyors would spread out at a defined distance (5 m, 10 m, etc.), walk over the terrain, and mark any archaeological feature encountered (site, structure, ceramic scatter, etc.). Full coverage surveys follow the same process, except they would cover every quadrat/transect in the region instead of only a selected assortment of them.

But these “traditional” survey techniques are not viable in the Petén Lowlands for several reasons. First, the dense jungles and lack of adequate roads and trails make traditional surveying a logistical nightmare. And the few trails that do exist follow old chiclero routes, which straddle high ground, while avoiding the low elevation, seasonal wetlands known as the bajos. Furthermore, the notorious Central American rainy season makes surveying difficult, if not impossible, from June to December each year. In fact, surveying is only practical during the dry season from March to May because the roads and trails require time to dry before they can be feasibly utilized.

Trails through the bajos can only be used during the dry season, and even then they are poorly suited for driving.

To overcome these challenges, Chiriboga has devised a new model that improves upon traditional approaches by incorporating a mix of traditional random survey techniques with data derived from remote sensing. This “hybrid” survey methodology aims to, in his words, “identify areas which would be preferred for long-term settlement based on a number of variables, not sites themselves, per se.” In other words, he is attempting to identify areas, in which the ancient Maya would likely have inhabited by looking at terrain characteristics, rather than blindly searching for ruins.

Strategically assessing the situation, Chiriboga has divided his study region along three main survey “strata” based on topographic prominence: the first stratum is comprised of the bajos, the second includes low-lying areas that do not flood seasonally, and the third stratum includes high-ground areas with relative topographic prominence. Due to their varying viability for long-term settlement, each stratum receives a specialized sampling methodology. Because past surveying has revealed a strong correlation between large sites and high topographical prominence, Chiriboga’s model gives areas in the third stratum a very high degree of survey intensity. In effect, the dry highlands receive sampling methodology similar to the traditional full-coverage survey, while the lowlands receive a random-sample survey.

Intuitively, this makes sense. Because highlands are at lower risk for flooding and have a host of other desirable char-acteristics, including greater distance from mosquito-infested wetlands, visibility of the surrounding terrain, and even a cool breeze, early Maya settlers would have been more likely to both settle and remain at a location in the highlands. In a sense, this is the logic of Chiriboga’s model – instead of randomly searching in the wilderness, it seeks to predict where the Maya would settle by thinking as they did.

Chiriboga utilizes the following data-intensive imaging components in his research models: NASA SRTM Digital Elevation Model (top left), Guatemalan National Institute of Geography’s 1:50,000 Topographic Series Maps (top right), NASA LANDSAT 5 T (bottom left), Final site prediction model with stratum one in blue, stratum two in green, and stratum three in yellow (bottom right). Photo courtesy of Carlos Chiriboga.

While simple in concept, this model is very data-intensive. Only after combining topographical maps, satellite remote sensing, digital elevation models, and a variety of other ecologic, hydrologic, and pedologic data can Chiriboga tease out the land cover information his model requires. To accomplish this, Chiriboga uses Esri’s ArcGIS software to integrate remote sensing data from NASA’s Landsat and Advanced Spaceborne Thermal Emission and Reflection Radiometer satellite sensors, complementing this information with the digital elevation models provided by the Shuttle Radar Topography Mission. In addition, Chiriboga has integrated maps from Guatemala’s National Geographic Institute, including their 1:50,000 scale topographic map series and their 1:250,000 scale geologic map. He has also used data from other archeological surveys in the neighboring lowlands to corroborate his model’s success in identifying areas of possible settlement.

After combining these sources of information, Chiriboga can then analyze the layers of spatial data to determine what stratum any given area is located within. For instance, Chiriboga identifies bajos by comparing elevation maps to seasonal moisture contents. Since bajos flood seasonally, they should have varying amounts of moisture depending on the time of year. In contrast, the areas of the second stratum would be relatively dry during the entirely of the year.

Chiriboga uses preliminary location models with surveyed sites represented by red triangles during his research.

Ultimately, Chiriboga’s model has proven to be an enormous success. After just two visits to the lowlands of Guatemala, Chiriboga has discovered 23 previously undiscovered sites, dating from the early Late Classic (c. 300 B.C.E.) to the Late Classic (c. 900 C.E.) Period. For now, he will continue to refine his model in the basement of 51 Hillhouse and work through his list of tetrads. Come March, however, Chiriboga will be gone, returning to explore the mysterious and rich jungles of the Petén Lowlands in search of ancient Maya ruins.

Pictured is a graphic representation of the layout of several of the sites Chiriboga has discovered.

About the Author
JOHN URWIN is a sophomore prospective Biology major in Jonathan Edwards College. He works in Professor Colón-Ramos’ lab, studying nervous system development in C. elegans.

Acknowledgements
The author would like to profoundly thank Carlos Chiriboga for his time, expertise, and passion for his research, as well as Jessica Schmerler for her assistance in researching this piece.

Further Reading
Puleston, Dennis, “Ancient Maya Settlement Patterns and Environment at Tikal, Guatemala: Implications for Subsistence Models” (January 1, 1973). Dissertations available from ProQuest. Paper AAI7414128

Rice, Prudence, The Kowoj: Identity, Migration, and Geopolitics in Late Postclassic Petén, Guatemala (University Press of Colorado, 2009)

Schwartz, Norman, Forest Society: A Social History of Petén, Guatemala (University of Pennsylvania Press, 1990)

Sever, Thomas, “Validating Prehistoric and Current Social Phenomena Upon the Landscape of the Petén, Guatemala,” Special Publication of the National Academy of Sciences/National Research Council on People and Pixels: Using Remotely Sensed Data in Social Science Research (National Academy Press, May 1998)