In a remarkable interdisciplinary breakthrough, researchers from Johannes Gutenberg University Mainz, the University of Oxford, and the University of Innsbruck have unraveled the intricate history of the ancient Roman aqueduct system that supplied the city of Arles in southern France. Leveraging an innovative approach centered on the analysis of aqueduct carbonates—essentially, the accumulations of limescale deposits found within the aqueducts and associated water infrastructures—scientists have reconstructed the evolution and operation of this sophisticated water management network that persisted for centuries. This groundbreaking study, published online on June 28, 2025, in the prestigious journal Geoarchaeology, offers new insights into Roman engineering brilliance and sustainable urban water management in antiquity.
The city of Arles, known for its rich Roman heritage and sprawling archaeological remains, was historically fed by a complex system of aqueducts, basins, and lead pipes. These elements worked seamlessly to ensure a steady supply of water, essential for the city’s baths, mills, and urban population. What sets this research apart from earlier studies is its holistic treatment of the entire water supply infrastructure, rather than focussing on a lone aqueduct. By investigating limescale deposits collected from multiple physical components of the system—including thick carbonate crusts from aqueduct channels and basin interiors, as well as carbonates embedded within architectural rubble—the team accessed an unprecedented archive of environmental and operational data.
Key to unlocking this historical narrative was the exacting study of what the researchers describe as “aqueduct carbonate archives.” Dr. Gül Sürmelihindi, a geoscientist with the Institute of Geosciences at Mainz University, emphasized that these carbonate deposits accumulate in successive layers over time, mirroring the operational history of the water system. “These sequential carbonate layers are akin to tree rings,” Sürmelihindi explained, “each encoding chemical signatures reflective of the water source, quality, flow characteristics, and changes in maintenance over decades and centuries.” Unlike previous work which often treated such deposits as mere nuisance or secondary artifacts, this study positioned them as primary historical witnesses capable of revealing Roman water engineering decisions.
Central to the Arles water system was an aqueduct originating around 3 BCE from the southern slopes of the Alpilles hills. This aqueduct was initially the sole conduit supplying the city, but after about ninety years, an additional aqueduct was constructed from the hills’ northern side. The waters of these two aqueducts merged in a strategically located header basin, which served as a sedimentation tank to trap suspended particulates—an early water purification step. The research team demonstrated that the northern aqueduct was an improvisational addition, reflected in its higher entrance level to the basin, indicating an adaptive evolution of the system in response to the city’s growing demands.
With the commissioning of the northern aqueduct, the southern aqueduct’s role was transformed. Instead of continuing solely to supply Arles, it was diverted to power the enormous Barbegal water mill complex, comprising sixteen large water wheels arranged in series down a hillside. This revelation reinforces previous hypotheses identified through carbonate analyses, highlighting the Romans’ deft capacity to integrate hydraulic energy solutions into urban supply systems. The Barbegal mill complex stands as a testament to the Romans’ pioneering use of renewable water power at an industrial scale.
Another piece of the puzzle involved the Baths of Constantine, monumental public baths constructed in the early fourth century AD under imperial patronage. These baths’ water source remained ambiguous until the researchers discovered lumps of aqueduct carbonate within the collapsed ceiling fragments of the baths’ structures. Isotopic and chemical analyses unequivocally linked these carbonate deposits to the northern aqueduct, thereby confirming it as the supply channel for the baths. Notably, the carbonate deposits were repurposed as building aggregate within the bath’s roof, offering a unique material record of aqueduct renovations contemporaneous with the bath’s construction.
This finding also allowed the researchers to date the aqueduct’s functional period well into the fifth century AD, challenging assumptions about the system’s decline and disuse in the late Roman Empire. The aqueduct evidently remained operational until the Frankish and Burgundian incursions disrupted the region’s institutions. This extended timeline underscores the durability and sustained maintenance of Roman hydraulic infrastructure even amidst the empire’s waning centuries.
The study also shed light on longstanding debates surrounding large Roman lead pipes discovered submerged beneath the Rhône River in the nineteenth century. There had been considerable uncertainty about the directionality and function of these pipes within Arles’ water network. By matching carbonates found inside the pipes with isotopic signatures from the northern and southern aqueduct carbonates, the researchers confirmed that these pipes formed part of an inverted siphon system, transporting water beneath the riverbed to supply the Trinquetaille district on the river’s opposite bank. This represents a sophisticated Roman application of hydraulic engineering principles, including the use of siphoning to overcome geographic barriers.
Isotope analysis constituted a pivotal methodological advance in this research. Due to heavy clay contamination, conventional dating methods were ineffective on the carbonate deposits. Instead, the team employed stable oxygen and carbon isotope profiling to establish depositional chronologies. These isotopic time series allowed for precise correlation of annual or multi-annual growth layers in distinct sections of the aqueduct carbonates, thereby reconstructing the temporal sequence of structural modifications, repairs, and water management strategies implemented over several centuries.
Professor Cees Passchier of Mainz University, a leading geoscientist who co-led the study, remarked on the power of this isotopic approach: “Without the aqueduct carbonate archives and the stable isotope methodology, it would have been impossible to disentangle the overlapping timelines and multifaceted hydraulic interactions we uncovered. This enabled us to illuminate one of the clearest examples of a sustainable water management system from antiquity.” The study thus not only deepens our understanding of Roman engineering prowess but also highlights ancient practices of infrastructure reuse, maintenance, and environmental adaptation.
The integrative nature of this research, combining archaeology, geoscience, chemistry, and engineering history, signals a promising direction for future studies of ancient infrastructure. Such analyses of mineral accretions embedded in hydraulic features have the potential to revolutionize how scholars reconstruct the chronology and functionality of large-scale urban water systems from numerous civilizations. This research model may soon be applied to other Roman aqueducts across Europe and beyond, opening new windows on the technological, social, and environmental dimensions of ancient city life.
Ultimately, this study spotlights the Romans’ profound understanding of water as both a vital resource and dynamic engineering challenge. The Arles aqueduct system’s multi-century operation exemplifies how ancient societies designed for resilience, adaptability, and sustainability—principles that retain relevance for contemporary water management. As climate change and urbanization pressure modern water infrastructures, revisiting the past through such detailed geoarchaeological investigations can yield valuable lessons on enduring stewardship of essential resources.
Subject of Research: Not applicable
Article Title: The Roman Water Management of Arles as Read in Aqueduct Carbonate Archives
News Publication Date: 28-Jun-2025
Web References: http://dx.doi.org/10.1002/gea.70020
References: Geoarchaeology journal, June 28, 2025 issue
Image Credits: photo/©: Cees Passchier
Keywords: Roman aqueduct, Arles, carbonate deposits, water management, isotope analysis, ancient engineering, hydraulic infrastructure, aqueduct history, Barbegal mills, Baths of Constantine, inverted siphon, sustainable water supply
