Presented at the first international EuroGPR conference, this paper asks a striking question: could ground-penetrating radar map the hidden interior of one of the world's most famous pyramids, without touching a single stone?
The subject
El Castillo, the stepped pyramid at the heart of Chichén Itzá, was built by the Maya during the Terminal Classic period (roughly 830–950 A.D.) and stands as a monumental representation of the Mayan calendar. It is one of the most visited archaeological sites in Mexico. While archaeologists have studied its artefacts and outward structure in detail, its internal properties have seen far less investigation.
What little is known hints at remarkable complexity. In 2015, researchers using electrical resistivity discovered a large cenote (a natural sinkhole) directly beneath the pyramid. In 2016, a third, roughly 10-metre pyramid was found nested inside two larger outer structures, like a Russian doll of Maya architecture. Beyond resistivity, however, non-destructive radar had barely been applied to pyramids at all. This work proposes to change that.
The proposed survey
The plan centres on an adapted Utsi Electronics GroundVue 7 GPR system fitted with a 40 MHz centre-frequency antenna. At that low frequency the antenna is a roughly 9-metre-long "snake", chosen to trade fine resolution for the depth needed to see through metres of solid limestone and fill.
- Non-metallic everything. Climbing equipment, hoists and ropes to move the antenna across the pyramid's stepped, non-horizontal faces must all be non-metallic, so they don't corrupt the radar image.
- Fibre-optic synchronisation. Transmitter and receiver sit on opposite sides of the pyramid and are time-synchronised over fibre-optic cable, avoiding the metallic cabling that would otherwise interfere with the survey.
- Total robotic station positioning. Every measurement is precisely georeferenced, because accurate position data is essential to reconstructing a clean image.
- Two-axis tomography. The structure is surveyed in both horizontal directions, moving transmitter and receiver in one-metre increments, to build a two-dimensional cross-section through the pyramid.
Modelling before digging in
Before any field survey, the paper builds simulation models in Sandmeier ReflexW to predict what the radar would actually see. A scale model of El Castillo, assembled from limestone and sand to match the expected materials and dimensions, is used to generate synthetic GPR data. The processing pipeline then runs in four steps:
- Build a representative scale model of El Castillo.
- Generate GPR data and store it against its position data.
- Extract the first-arrival "travel time" from each transmitter–receiver measurement.
- Reconstruct a tomography image using the Simultaneous Iterative Reconstruction Technique (SIRT).
A second, more demanding model adds the pyramid's stairs and a 10 m² internal void, deliberately echoing the hidden chambers found in the Egyptian pyramids, to test whether radar could reveal such a feature.
What the modelling showed
- The boundaries between the three nested structures appear clearly in the tomography images as changes in contrast, suggesting radar could distinguish El Castillo's internal layers.
- Precise positioning and good transmitter–receiver geometry are fundamental to obtaining accurate images.
- Material properties ultimately govern how deeply the radar can penetrate: the physics of the structure sets the limit.
Why it matters to Norse Radar
This is subsurface imaging at its most ambitious: reading the inside of an irreplaceable monument entirely non-destructively. The same principles (careful antenna choice, rigorous positioning, disciplined processing and honest modelling of what radar can and cannot see) underpin every Norse Radar survey, whether the target is a Maya pyramid or the ground beneath a construction site.
Acknowledgements to Vincent Utsi (Utsi Electronics) for GPR expertise and Adam Szynkiewicz (University of Wrocław) for research guidance.
