Rescuing the Bayon: Three Decades of Discovery & Defiance

When you stand before Bayon Temple’s mysterious and enchanting stone faces today, what you’re witnessing is the result of one of the most sustained and scientifically rigorous conservation efforts in Southeast Asian archaeology. Since 1994, the Japanese Government Team for Safeguarding Angkor (JSA/JASA), working alongside UNESCO and Cambodia’s APSARA National Authority (ANA), has been methodically preserving, researching, and revealing the secrets of this extraordinary temple.

It is a story of defiance against the ravages of time, centered on an iconic tower at critical risk of collapse. The project evolved into a detective story where each season of work uncovers new mysteries, where traditional Khmer construction techniques are rediscovered, and where the foundations of what we thought we knew continue to shift beneath our feet.

The Monument and the Challenge

Bayon, constructed in the late 12th to early 13th century during the reign of King Jayavarman VII, sits at the exact centre of Angkor Thom. With its 54 towers bearing 216 serene stone faces (173 still existing), it represents the architectural and spiritual apex of the Khmer Empire in the Jayavarman VII era. But it also represents something else: one of the most technically challenging conservation projects you could imagine, rivalling that which took place at the nearby Baphuon.

Unlike the precisely planned Angkor Wat, the Bayon was built using what scholars describe as ‘responsive field techniques,’ implying construction decisions made on-site rather than from predetermined plans. This created a high-density cluster of structures with unique engineering challenges. Add to that centuries of tree root damage, irregular foundation subsidence, and the circular plan of the Central Tower (unprecedented in Khmer architecture), and you begin to understand the scale of the undertaking.

A Journey Through The Six Phases of the Bayon’s Restoration

A Timeline of Conservation, Research, and Archaeological Revelation 1994-2026: It’s a story that unfolds across six distinct phases, each building upon the discoveries and technical breakthroughs of its predecessors. Let’s walk through them.

Phase I (1995-1999) – Finding the Foundation

The work began with Bayon’s North Library, which had started to collapse in 1995. This wasn’t a case of “let’s start here and work our way around”, nor about fixing a single building. This was about establishing the very methodology that would guide everything that followed for decades to come.

Here’s where things get interesting. Previous restoration efforts at Angkor had often replaced traditional construction methods with modern materials, in the process ripping out the ancient rammed earth cores and replacing them with reinforced concrete. Certainly practical when you’re short on time and materials, but you lose something irreplaceable: the original structural logic, the traditional load-bearing systems, the very essence of how Khmer engineers understood their craft.

The JSA team took a different approach. They developed a lime-stabilised rammed earth technique that mimicked the original construction while providing modern structural enhancement. The method works like this: soil is mixed with minimal lime additive, carefully matched to the original particle size distribution (determined through laboratory analysis), then compacted using traditional methods. The result maintains the original permeability, which is apparently crucial for managing water movement, while achieving the engineering standards required.

But there was another problem. Where do you source replacement sandstone that matches the original material? The French had identified quarries decades earlier, but by the mid-1990s, these areas were either inaccessible due to landmines and political instability, or simply forgotten. After two years of joint exploration with Cambodia’s Ministry of Industry, Mines and Energy, JSA successfully located matching sandstone at Phnom Kulen, 30-40km northeast of Angkor.

This single breakthrough of sourcing original-quality materials opened the door for authentic conservation across the entire Angkor region.

Phase II (1999-2005) – The Masterplan Takes Shape

If Phase I was about establishing methods, Phase II was about establishing vision. The work expanded beyond Bayon to include Prasat Suor Prat towers and Angkor Wat’s North Library, but the real achievement was the publication of the Bayon Master Plan in 2005.

This wasn’t some dusty academic document. The Master Plan represented the culmination of nine annual symposia (1996-2004) where international teams, Cambodian scholars, and conservation experts hammered out a comprehensive framework for the temple’s preservation. Think of it as a constitution for Bayon’s conservation, in turn establishing principles that balanced scholarly research, traditional techniques, international cooperation, and the competing demands of tourism, national heritage, and World Heritage obligations.

On the practical front, Phase II saw significant technical achievements. At the nearby Prasat Suor Prat, the team conducted the first comprehensive laterite structure restoration with full foundation excavation, including disassembling approximately 2,000 laterite blocks and 700 sandstone blocks. At Angkor Wat’s North Library, they pioneered a complete digital documentation system, creating HTML-based databases for all construction records, drawings, and repair documentation.

It’s worth noting the philosophy behind the restoration approach. When reconstructing collapsed sections, new carved stones are finished to match the original carving technique, then intentionally ‘aged’ using bush hammers and chisels. The result? New work that fits in visually while remaining identifiable upon close inspection to satisfy both conservation ethics and aesthetic integrity.

Phase III (2005-2011) – South Library and Sacred Discoveries

With $3.27 million in funding, Phase III focused on Bayon’s South Library while simultaneously launching crucial research into the Central Tower’s stabilization and the preservation of the inner gallery bas-reliefs.

The South Library work marked a shift in approach. While Japanese experts remained involved, Cambodian specialists took the lead. The structure, similar in form to the North Library, required disassembly beginning in 2006, with reconstruction completed in 2011.

But the real excitement came from archaeological discoveries during the disassembly. Inside the platform core, the team found consecration deposits (ritual objects placed during the original construction) along with nested internal structures that revealed how the building was conceived and built. These discoveries provided crucial evidence for understanding Khmer construction ritual practices and chronological sequencing.

Meanwhile, research teams were deploying an arsenal of investigative techniques on the Central Tower: structural monitoring, microtremor measurements (to understand seismic response), climate data collection, groundwater tracking, and comprehensive ground surveys. The picture that emerged was sobering. The 42-meter tower had significant structural vulnerabilities, particularly in its foundation.

And then there were the bas-reliefs. The approximately 300 meters of shallow carved reliefs in the inner gallery were deteriorating at an alarming rate. A multidisciplinary team combining petrology (stone science), microbiology (studying the organisms attacking the stone), and conservation science began developing preservation strategies. Their findings would become crucial for the work ahead.

Drainage systems were also comprehensively studied, in this case by Sokuntheary So et al., confirming the complex drainage system and its function. Not only categorising them by having a sacred function or being purely functional, but also chronologically.

Phase IV (2011-2018) – Eastern Façade and Landscape

Phase IV shifted focus to Bayon’s eastern approach, especially important as it’s the face that most visitors first encounter. Tower 55 and the eastern entrance gopura (gate) were both heavily deformed, their stones displaced by centuries of structural movement.

While restoring these structures, archaeological investigations revealed something unexpected about the eastern ponds. Previous scholarship had assumed a single pond configuration, but excavations uncovered evidence of two distinct water features on the eastern side. More on that later.

Phase IV also saw cooperative projects, including the restoration of Naga balustrades and lion sculptures along the outer gallery and eastern causeway. These weren’t JSA’s primary responsibility, but the team provided technical support, which was another example of the knowledge transfer that runs through this entire project. Importantly, investigations into methods for the permanent stabilisation of Bayon’s Central Tower continued.

On the bas-relief front, the team implemented experimental treatments by testing different cleaning methods, consolidants, and waterproofing techniques for the roof joints. Each experiment was carefully monitored, the results feeding into an evolving preservation protocol.

Phase V (2018-2020) – Crisis and Closure

February 1, 2020. The Central Tower closes to visitors. After years of investigation, the structural analysis had identified extremely fragile sections in the foundation directly beneath the tower. These weren’t minor concerns as they represented genuine collapse risk for one of Angkor’s most iconic structures.

With $1.5 million in funding, Phase V focused on developing implementation strategies for the critical foundation reinforcement. The challenge was unique: how do you stabilise the base of a 42-meter stone tower without dismantling it entirely? The circular plan, unlike anything else in Khmer architecture, meant there were no direct precedents to follow. After extensive research, Central Tower reinforcement would focus on soil stabilisation and water management.

At this time JASA also provided technical assistance to APSARA for work on the Third Terrace, where Cambodian specialists were increasingly taking the lead on restoration operations. This handoff with Japanese experts transitioning to advisory roles while Cambodian teams take charge represents the ultimate goal of the entire 30-year program.

Phase VI (2020-Present) – The Current Chapter

November 2020 brought the start of Phase VI, initially supported by $520,847 for equipment and machinery. Then, in December 2024, Japan committed an additional $900,000 to carry the work through 2028.

Recent Work and Discoveries (2020-2025) – The Details

While the phases give us the big picture, the real story emerges in the details of recent work. Let’s zoom in on what’s been happening year by year.

2021 – Stone by Stone

Early 2021 saw APSARA begin work on the southwest corner of the third terrace, with JSA providing technical guidance. By June, teams were systematically relocating nearly 3,000 ancient stones from the moat area that included fragments from the French restoration campaign (École française d’Extrême-Orient, 1913-1929) and pieces from the temple’s original structural collapse.

Think about that for a moment. Every one of those 3,000 stones had to be documented, photographed, measured. Each might hold clues to where it originally belonged, how it was carved, when it fell. It’s a massive three-dimensional jigsaw puzzle where most of the pieces are missing and there’s no picture on the box.

By August, work at the southwest corner of the third terrace (uppermost tier) wrapped up with six months of entirely manual work without heavy machinery. Less than 20% of stones were replaced, with new sandstone carefully sourced from the same Phnom Kulen quarries identified back in Phase I.

November 25 brought good news for visitors: most temple areas reopened with newly designed touring routes. The routes balance conservation needs with visitor access while keeping people away from the most vulnerable areas, yet still allowing them to experience the temple’s grandeur. The Central Tower, of course, remained closed.

2023 – The Two-Story Discovery

September 18, 2023 brought one of those moments that must make archaeology exciting.

While working on the third terrace, the restoration team uncovered a two-story overlapping platform structure consisting of two separate platforms, built at different times, one on top of the other. The older lower platform showed simpler decorative elements compared to the upper platform’s more elaborate ornamentation.

This is significant. It suggests that even during Jayavarman VII’s reign, the design intent or religious program evolved. The temple was not built according to a fixed plan; instead, it was modified in situ, evolving for reasons that remain a mystery. This kind of overlapping construction is fundamentally different from what is seen at other Angkorian temples, where construction phases are typically distinct and sequential rather than superimposed.

It was the second such discovery at Bayon, suggesting this might have been a deliberate architectural approach rather than a one-off modification.

2024 – The Year of Water and Stone

January brought announcements about the southeastern pond restoration. Excavation work revealed sediment layers that could be analyzed chronologically (think of it as reading the pond’s history in reverse) along with features of the original pond construction.

By late January, the latest southwest platform third terrace campaign wrapped up after four months. The work included precise repositioning of ridge stones, compaction of the drainage system foundation using lime-stabilized soil (that Phase I innovation still paying dividends), and complete restoration of a 12-meter Naga balustrade.

January 31 saw the start of work on the ancient drainage system in the first gallery southwestern sector. This meant systematically removing and documenting hundreds of stones originally placed by French conservators in the early 20th century. The French, working with limited resources and under time pressure, had sometimes blocked or modified original drainage features. Now, armed with modern archaeological methods, teams could restore the original Khmer hydraulic engineering.

Then came April’s surprise.

At 1.5 meters depth in the eastern pond, excavators found a beautifully preserved sandstone turtle sculpture measuring 57cm × 43cm × 21cm. In Hindu cosmology, the turtle is fundamental to creation myths, and perhaps had other virtues in Khmer mythology, with many found appearing to be sacred deposits.

The turtle’s discovery confirmed what the excavations had been suggesting (as noted earlier): Bayon had two distinct ponds on its eastern side, not one as previously assumed. This fundamentally revised the understanding of the temple’s ceremonial landscape. By April 22, the sculpture was carefully transported to Preah Norodom Sihanouk-Angkor Museum for conservation treatment and analysis.

July saw intensive work removing approximately 4,000 stone blocks from the eastern ponds, which sounds like a lot. The estimated total for both ponds? Around 20,000 blocks! Each one needs documentation using photogrammetry and measured drawings, which is hard to imagine. The assemblage includes roof gallery components, structural beams, column elements, and platform stones. Some might be reassembled, others will inform research, and be catalogued.

December brought validation and commitment. On the 8th, an ad-hoc expert panel conducted comprehensive inspections of restoration sites at West Mebon and Bayon, reviewing methodology and structural integrity. Two days later, the 30th anniversary symposium. And on December 13, Japan’s announcement of $900,000 for the 2025-2028 project ensured continuity for the critical work ahead.

The year closed on December 19 with the completion of the third terrace southwest corner restoration campaign. Four months of work restoring deteriorated Naga balustrades, structural beams, and stone flooring using traditional anastylosis combined with modern stabilization techniques.

2025 – Looking Forward

May 8 saw APSARA initiate a comprehensive stone registration program for architectural elements dispersed around Bayon’s precinct which is an important precursor to further restoration work.

In June 2025, another surprise was the discovery of a torso that could be an Avolokiteshvara statue. Missing the head and feet, it may have originally been somewhere around 1.5m tall.

By August, new access stairs were installed for the third level, improving visitor circuit and safety while minimising impact on original stonework.

Conservation Methodology & How They Actually Do It

Now, here’s where we get into the fascinating technical details. Because saying ‘they restored the temple’ leaves out the entire story of how.

The Philosophy

JSA/JASA’s approach rests on five principles: maximum retention of original fabric and construction techniques; structural reinforcement that respects original engineering; minimal intervention that’s reversible when possible; complete documentation of all work; and transfer of knowledge to Cambodian specialists.

That last point deserves emphasis. The ultimate measure of this project’s success isn’t the restored stones, it’s whether Cambodia has the expertise to manage its own heritage when the Japanese teams eventually step back.

Traditional Methods Meet Modern Science

Take the manual anastylosis methodology. This is a complete disassembly and reconstruction following the original construction sequence. Each stone gets cleaned, documented, and repaired if necessary. Then it’s repositioned with millimetre precision. In sensitive areas, the teams use hand tools exclusively without machinery that might cause mechanical damage.

Or consider the ‘Elephant’s Foot’ compacting tool. This is a traditional Khmer foundation compaction implement, basically a heavy-weight dropped repeatedly to compact soil. The teams use it for platform and terrace substructure work, replicating original methodology while achieving modern engineering standards. It’s slower than mechanical compaction, but it maintains the soil structure in ways that machines can’t match.

Then there’s the lime-stabilised rammed earth technique we mentioned earlier. The innovation here is in the balance: minimum lime addition mixed with traditional rammed earth construction. The soil mixture is carefully matched to the original particle size distribution (determined through laboratory analysis). The result maintains original permeability, which is crucial for managing water movement through the structure, while providing significant strength enhancement.

But some things require modern intervention. Stainless steel dowels with epoxy resin (mixed with sandstone powder for color matching) get used for critical structural repairs. Polymer-cement mortar, colored with sandstone powder and inorganic pigments, protects exposed joints from weathering. Mobile cranes and mini-crane trucks ensure worker safety during heavy lifting. The art is knowing when traditional methods suffice and when modern support is necessary.

The Research Arsenal

Perhaps the most fascinating aspect is the sheer breadth of investigative techniques deployed.

Magnetic susceptibility analysis measures sandstone magnetic properties without damaging the stone. This enables identification of quarry sources, construction chronology, and modification episodes. Using this technique, researchers successfully dated construction from the Angkor Wat through Bayon period and identified reused stones from earlier structures.

The structural monitoring network provides continuous data: displacement monitoring tracks how stones are moving, microtremor measurement reveals seismic response, climate data (temperature, humidity, wind, rainfall) identifies environmental stresses, and groundwater level tracking shows how the water table affects stability. All this information feeds into conservation decisions and validates whether structural interventions are working.
3D documentation through photogrammetry and laser scanning creates comprehensive digital records. More than just fancy pictures, they enable damage assessment, structural analysis, virtual reconstruction experiments, and create a permanent archive independent of the physical structure’s condition.

Finally, multidisciplinary deterioration studies integrate petrology (stone characterisation, weathering mechanisms), microbiology (biofilm identification, biological deterioration), conservation science (cleaning methods, consolidant testing), and environmental monitoring. This integrated approach is particularly crucial for the endangered bas-reliefs, where surface deterioration threatens to erase irreplaceable historical narratives.

The Persistent Challenges

The biological siege is relentless. Roots wedge into stone joints, bat guano provides an acidic threat, and a cocktail of bacteria, fungi, and lichens erodes the surface. Compounding this, chronic water leaks and failed drainage could undermine the temple from within.

If you are an irrigation farmer, you’ll know how insidiously hydraulic deterioration works. As a kid, I watched my father fight against it for years. In the Bayon’s case, rainwater infiltrates through roof joints, causing interior erosion while capillary moisture movement concentrates salts at stone surfaces, and ancient drainage systems, compromised over centuries, will require constant investigation.

What does seem apparent at the beginning of this multi-phase project is that foundation instability posed the greatest long-term risk. The massive stone superstructure rests on soil that compresses unevenly with water table fluctuation, causing seasonal ground movement. As confirmed in 2009 in a study by Shimoda Ichita, one of the critical weak zones lies directly beneath the 42-meter Central Tower. The incredible tower features a central shaft, where Georges Trouve discovered the “Bayon Buddha” in 1933 and was later backfilled. To some surprise, this 42m tall tower was only supported by compacted sand and laterite chips.

And that is the core of what we have been watching over the past decades, not just restoring the beautiful facade, but also solidifying, sealing, and restoring functionality that we may not even see and appreciate beyond the placement of new stone masonry.

Where Things Stand Today

Perhaps the most significant achievement of this project isn’t visible at all. It’s the establishment of a sustainable conservation methodology applicable throughout Angkor Archaeological Park. The training of Cambodian specialists in traditional construction techniques, modern conservation science, and project management ensures long-term stewardship by the nation’s own experts. That’s the program’s most enduring legacy.

As of February 2026, Phase VI continues with renewed Japanese commitment through 2028. APSARA increasingly assumes primary conservation responsibility, with JSA/JASA transitioning to technical advisory roles. This handoff, which I assume is where things are now, fulfils the original capacity-building mandate.

The project priorities of implementing the foundation reinforcement strategy for the Central Tower, bas-relief preservation program, northeast corner building restoration, are reaching a conclusion while new funding will support repairs to ancient drainage systems and aid in the registration of scattered stones around Bayon Temple. APSARA also plan to reopen the conservation building on the south side of the temple for stone registration and, as had done previously, provide deep insight into the construction and bas-reliefs of the temple .

Ongoing Mystery

Here’s what strikes me most about this 30-year effort: the more we all learn, the more we all realise how much we don’t know. That is the story of ancient Angkor, after 100 years of research, we are still peering through a window that’s only fractionally open.

The two-story platform discovery changes our understanding of Bayon’s construction sequence. The turtle sculpture rewrites the temple’s hydraulic landscape. Each season of work reveals new architectural details, new construction techniques, and new questions about its early architectural development, how Jayavarman VII’s vision evolved, and, in turn, how it evolved once again in the late/post-Angkor era.

The restoration of a replica Bayon Buddha to the main chamber of the central tower remains a mystery at this time, as to whether it’s going ahead or not. The replica sits in storage on the north side of the Bayon while the original remains at Preah Vihear Pram Pi Lven (or Terrace Sisowath). Let’s wait and see for that one. If it goes ahead, it’s a big deal, but also conflicting. Who is then the pivotal ancestral Buddha for the local Khmer?

In many ways, the Bayon isn’t just being preserved; it’s being continuously revealed. And that, perhaps, is exactly as it should be, as great monuments shouldn’t yield all their secrets at once. They should remain mysteries that each generation gets to explore anew, armed with better tools and fresh perspectives. We are all just students at its feet, and to be imparted with some of that knowledge is a blessing for sure.

The stone faces will keep their counsel for now, watching as scholars and craftspeople work to understand and preserve their entrancing smiles for generations yet to come.

Sources and Further Reading

Primary documentation comes from the Japanese Government Team for Safeguarding Angkor (angkor-jsa.org), APSARA National Authority announcements, UNESCO reports, the Bayon Master Plan (2005), and Bayon Charter (2004).

For those interested in diving deeper:

• Our Visiting Guide and Introduction to the Bayon Temple
• the JSA/JASA website maintains comprehensive technical reports across all phases
• the APSARA Authority provides updates on current work.
• ICC Angkor Technical Committee Reports offer scholarly perspectives on specific conservation challenges and discoveries.