Image - Cacao Pod Vessel - K6706 © Justin Kerr FAMSI © 2003:
Francisco Estrada-Belli
(Vanderbilt University)
 

Archaeological Investigations at Holmul, Petén, Guatemala
Preliminary Results of the Third Season, 2002

Discovery of Hahakab

The discovery of Hahakab (Yucatec Maya for "looking at the earth"), a minor center in the vicinity of Holmul, was made possible by GIS analysis of digitized maps, terrain modeling, viewshed and optimal path models developed at Vanderbilt University with the assistance of anthropology major Christine Fitzgerald (class of 2001). The steps used for building the site’s prediction model can be summarized as follows:

  1. A terrain surface model was created from the Guatemala IGM 1:50000 maps covering an area within a 15 km radius from Holmul (Figure 5). The maps’ elevation intervals of 20 m permitted the modeling of the complex landscape of rolling hills, escarpments and swampy flats around Holmul.
  1. An attrition map (cost surface) was developed by combining changes in elevation, slope, and slope direction as factors to facilitate or impede movement through the landscape. A separate attrition surface was created by adding a map of the viewshed area from the vantage point of Holmul’s tallest pyramid to the topographic factors. According to this model, areas within view of Holmul Group I pyramid were considered easier to traverse than areas outside its view. The rationale for using viewshed as a factor for access routes was that traveling within view of the destination would have provided ritual significance to the path and as well as defensive properties for the travelers and Holmul residents.
  1. Points were selected at the extremes of the area of interest in connection with the four cardinal directions. From these locations, movement was to be modeled into Holmul by following the paths of least-cost through the attrition surfaces. The northern point was chosen to simulate the hypothetical path someone traveling from the city of Xmacabatún (30 km N) would have taken to approach Holmul. The western point simulated travel from the general direction of Tikal and Uaxactún through the Bajo de Azucar. The eastern point represented hypothetical routes originating at an hypothetical location below the escarpment, north and east of El Pilar (Lamanai?). The southern point represented the approximate location of Naranjo and thus simulated travel from there to Holmul.
  1. The GIS path algorithm produced two sets of paths leading to Holmul from each of the outer point locations. One set included four optimal paths through the topography only while the second set represented four paths through the topography with the addition of the viewshed variable.
  1. The results were compared to the topographic map and location of existing minor centers situated within a 3-4 km distance of Holmul (Figure 5). In both sets, paths from the west were found to be approaching Holmul from the vicinity of the minor center of T’ot. Both paths from the south approached Holmul passing directly through the location of the minor center of Riverona. One eastern path passed very close to the K’o center, while a second was later found to pass through the newly discovered site of Hamontun (see "Other Explorations" below). One should note that each of the southern peripheral centers were also located at the edge of the Holmul upland area facing the swampy flats. The northern route however passed through an area which had never been explored. It was noted that the two northern paths (viewshed and non) crossed the northern swamp along separate routes but joined on a common course once they approached the upland plateau (closer to Holmul) and came within the viewshed of Holmul, as well. This location at the edge of the Holmul viewshed and along the two joining paths was selected as a possible location for a northwestern center that would have completed the circle of centers around Holmul at a distance of 3-4 km.
  1. Landsat imagery and a Star3i high-resolution radar image (provided to Dr. Magaly Koch, courtesy of NASA) were analyzed visually in search of possible indications of major architecture buried under the thick forest cover. The target location suggested by the GIS model was confirmed by concentration of tree species consistent with upland terrain and tall rocky outcrops visible on the imagery, but deemed undistinguishable from possible Maya buildings.
  1. The coordinates obtained from the GIS maps were entered into a hand-held GPS receiver which guided the survey team composed of Justin Ebersole, Britta Watters, and Shoshuanna Parks to the target/test location under a thick canopy vegetation. The ceremonial center of Hahakab was located within 100 m of the predicted location.

The main ceremonial area of Hahakab was mapped by Justin Ebersole using tape and Suunto compass (Figure 6). The site is composed of about 20 major structures, one of which, the tallest, measures 18 m in height. The main group of the site is formed by a triadic pattern of small pyramids on an elevated platform. This group also includes other buildings on the front of the stairway and at the NW and NE corner of the platform (see Uaxactún Group H pattern, Valdes 1989). To the south of this platform is a plaza and E-Group complex. To the east of the E-Group and triadic group is an elevated platform supporting structures of possible elite residential function. To the north this platform and the Triadic Group face an open plaza enclosed on the east by a tall pyramid (Str. 1). Axial looters’ trenches in this structure bisected several earlier stages of construction.

Within the plaza of the Hahakab complex, three plain monuments were found resting on the humus surface.

Previous Page  |  Table of Contents  |  Next Page

Return to top of page