Since 2001, a number of ground-penetrating radar (GPR) surveys, and near-surface seismic surveys have been conducted at the ancient Maya site of Ma'ax Na in Belize, Central America. The surveys were initiated to assist archaeologists in focusing their excavation activities.
The Ma'ax Na project involved the surveying of several sites within the complex including the plaza and several caves. GPR acquisition consisted of several orthogonal 2-D lines, and two 3-D grids. The report will deal with the GPR records of the plaza only.
This research project has focussed on improving the quality of the GPR images generated from the Ma'ax Na plaza lines, and has attempted to interpret these images based on the archaeological information. In doing so, a more definitive processing and interpretation flow has been established.
The survey was acquired using Sensors and Software's Noggin® and Smart Cart® System, at a GPR antennae frequency of 250 MHz. Spatial sampling of 5 centimetres and a temporal sampling of 0.04 nanoseconds provided reasonable quality records, with good signal penetration. The maximum penetration depth of the plaza lines was about 1.8 m.
The recorded velocities in 2002 ranged from 0.106 m/ns @ 0.7 m and 0.072 m/ns @ 1m and were calculated from the hyperbolic fitting of curves to point diffractors. In 2003, the measured velocities were noticeably higher in the range of 0.122 to 0.140 m/ns. We attribute this increase due to very dry conditions in the area compared to the year before.
Resolution based on a velocity of 0.072m/ns and the antennae frequency of 250 MHz was approximately 7.2 cm. Excellent ties between intersecting plaza lines and the duplicity between two lines shot in reverse directions inspired confidence in the accuracy of the acquisition. Modelling of the GPR based on changes in the electrical properties of dielectric permittivity revealed a reasonable correlation between the synthetic radargram and the GPR image. Surficial discontinuities (roots) and anomalous buried features were identified.
View full article as PDF (6.74 Mb)