Building a next-generation gravitational-wave observatory means digging a lot of rock, and a new study argues that what happens to that rock could decide whether the project is ever built at all. Researchers D. Schneider and R. Galler of Montanuniversität Leoben have published a framework for managing excavated material in the early planning phases of large underground research infrastructures, applied as a case study to the Einstein Telescope (ET), Europe’s proposed third-generation gravitational-wave observatory. Their central argument is that the storage, transport, processing and utilisation of tunnel spoil fundamentally affects project approvability, environmental impact, total cost and public acceptance, and that these dimensions must therefore be systematically managed under uncertainty from the earliest possible project stage.
The scale of the challenge is considerable. The Einstein Telescope will consist of underground caverns and tunnels roughly 10 to 15 kilometres in length, depending on the geometry variant ultimately chosen, with an external diameter of more than seven metres at a depth of around 250 metres. The estimated total excavated volume lies between 3.5 and 4.4 million cubic metres in situ, a figure expected to be updated by autumn 2026, corresponding to up to 5.3 million cubic metres after bulking with a factor of 1.2. In hard rock conditions, even larger bulk factors are expected. A significant proportion of this material arises within a comparatively short time window during tunnelling operations, and past tunnelling projects show that direct project-internal use will necessarily be limited, since on-site demand for backfill, support and ancillary works typically represents only a fraction of the total volume generated by infrastructure of this scale.
The paper, published as a review in EPJ Research Infrastructures, responds to the context of the Einstein Telescope Preparatory Phase (ET-PP), the project stage in which foundations are laid for turning the ambition of an underground observatory into a realistically plannable and approvable major undertaking. The ET was recognised as a priority European research infrastructure by an ESFRI decision on 30 June 2021, and the Einstein Telescope Collaboration, formally constituted in 2022, now associates approximately 1,900 researchers from 271 institutes in 31 countries as of 2026. Two observatory geometries remain under discussion: a single-site triangular configuration with three interlocked detectors, and a variant with two L-shaped interferometers. Because the geometry decision has not been reached, the authors stress that any material strategy must be built on a scenario basis comparing at minimum both layouts.
Methodologically, the framework follows a process- and criteria-based approach built on four coupled elements. The overall project context establishes system boundaries; material management is modelled as a governance problem under uncertainty, in which risk maturity rather than planning completeness determines when utilisation decisions become binding; quality management is treated as a continuous process chain; and utilisation planning is framed as a portfolio-matching problem, mapping material properties, quantities and temporal availability against use-case requirements and site-specific constraints. This last element distinguishes the approach from conventional spoil logistics planning, which typically relies on laboratory measurements of material after excavation rather than on probabilistic characterisations derived from site investigation ahead of construction, combined with online characterisation directly during excavation.
The risk management component follows ISO 31000 and proceeds through sequential steps of risk identification, qualitative analysis, quantitative analysis and measures definition, with continuous monitoring running in parallel. Each risk is described with event and impact, assigned cause, pre-classification by type, location and rough criticality. Qualitative analysis prioritises risks through monetary assessment, probability estimation and temporal relevance, while quantitative analysis yields numerical contributions to cost and schedule reserves. For the ET-PP context, the authors identify risks to be addressed early, including geotechnical and hydrogeological uncertainties, fluctuations in excavated material quantity and quality, approvals and cross-border compliance, logistical interfaces to reuse and disposal markets, supply chain and offtake market capacities, and social acceptance.
The proposed Quality Management Strategy defines material segmentation into geologically and operationally homogeneous batches, each assigned unique georeferenced identifiers and linked to a regulatory classification using EU List of Waste codes. Identification at the tunnel face follows the ISO geotechnical description framework, using ISO 14688 for soils and ISO 14689 for rock. Minimum parameters include particle-size distribution and fines content, moisture and density, mechanical resistance such as compressive strength and the Los Angeles coefficient, frost susceptibility and compaction behaviour, with additional chemical parameters for sensitive applications and leaching tests where water contact is anticipated. Where analytical results exceed acceptance limits, a controlled non-conformance workflow places the affected batch on hold, verifies triggering measurements through re-testing by an ISO 17025-accredited laboratory, conducts root-cause analysis, and follows a decision hierarchy of reworking, reassignment to a lower-grade use, or classification as waste with licensed disposal. Notably, because the candidate sites span divergent national regulatory regimes operating in parallel on a single excavation programme, the strategy is designed around a jurisdiction-neutral traceability backbone rather than any single national template.
The case study examines the three official candidate regions: the cross-border Euregio Meuse-Rhine (NL/BE/DE), Sos Enattos in Sardinia, and Lusatia in Saxony, the latter added as third candidate in October 2025. At Sos Enattos, near a former mine, the geology is dominated by crystalline hard rock of the Variscan basement, intruded by leucogranites and granodiorites; core drillings to approximately 270 metres depth led to an estimated Rock Mass Rating of 20 to 40, classified as poor quality. The Euregio Meuse-Rhine features Upper Devonian sandstone of rather high strength overlain by Lower Carboniferous clay- and siltstone and Cretaceous chalk, with variable to locally high hydraulic permeabilities implying sectionally relevant water ingress at 250 metres depth. Lusatia features the Lusatian granodiorite, a massive intrusive rock in which a borehole to 250 metres depth confirms high strength and low permeability below the weathered surface zone. Using an Excel-based screening model developed at Montanuniversität Leoben, the authors find that EMR sandstones with moderate strength above 100 megapascals are feasible as recycled building materials, though slightly elevated arsenic and cobalt values require attention for leaching-sensitive applications, while the granodiorites of Lusatia and Sos Enattos are very well suited to high-quality aggregates, subject to grain characteristics from tunnelling and EN 12620 requirements. Micaschist at Sos Enattos, at approximately 60 megapascals, presents challenges from platy grains along schistosity planes and mica-rich fine fractions. The authors deliberately omit a comparative quantification across the three regions, since the site decision is outstanding and detailed rankings could influence the ongoing selection process.
On the legal side, the paper concludes that all frameworks relevant to the three candidate sites already enable high-quality circular economy use of excavated materials, though via different pathways: the Netherlands emphasises certified soil-quality regulations and integrated permitting, Flanders combines material policy instruments with traceability, the German EBV provides uniform federal rules for mineral secondary raw materials, and Italian decrees DPR 120/2017 and DM 127/2024 enable by-product status and end-of-waste markets. The recommendation is to design material management on on-site reuse hierarchies, integrate digital building logbooks from the early design phase, clarify end-of-waste pathways with regional authorities in advance, and organise cross-border waste streams under the Waste Shipment Regulation. The authors also highlight innovative pathways, including alkali-activated materials and geopolymers for clay-rich material, thermally activated clays as supplementary cementitious materials, and agricultural topsoil engineering combining sterile mineral content with organic residues in a 7:3 ratio.
A forward-looking element is sensor-based online characterisation of material streams. The proposed system combines Raman spectroscopy for mineralogical analysis, laser-induced breakdown spectroscopy (LIBS) for chemical and elemental composition, and hyperspectral imaging for grain size, shape, volume flow and chemical signature, supplemented by belt scale weight and moisture measurement. Sensor data would be linked with prior geological information, and an AI-supported decision matrix would assign material flows in near real time to defined utilisation pathways. The authors note that the technology readiness level of the combined system for industrial-scale tunnel applications is currently TRL 4 and remains to be validated through planned full-scale trials, with a measurement bridge in development for an underground research facility with an industrial-scale conveyor belt system. If successfully validated, the transition from batch-based retrospective classification to continuous, predictive material flow management could significantly reduce interim storage requirements and improve the economics of circular material utilisation.
The framework’s relevance extends beyond the Einstein Telescope. The authors compare it with the Future Circular Collider at CERN, whose Feasibility Study Report was completed on 31 March 2025 and which foresees an underground ring of approximately 91 kilometres circumference with excavation estimated in the millions of cubic metres of molasse from the Geneva Basin. Both projects share the defining condition that utilisation governance must begin before geometry, construction method and host arrangements are settled, and across jurisdictional boundaries that no single national template fully governs. The authors caution that what transfers between infrastructures is the governance logic and characterisation methodology, not site-specific parameters or quantitative results. In all cases, they conclude, the present data foundation is insufficient for definitive planning decisions and must be expanded through targeted site investigation, with reports from ongoing campaigns expected to be published by the end of 2026 and recommendations on the site and geometry decision projected for 2026 and 2027.
Subject of Research: A risk and quality management framework for excavated material from the Einstein Telescope underground observatory
Article Title: Excavated Material Strategy for Large Underground Research Infrastructures in Early Planning Phases: A Risk and Quality Management Framework Applied to the Einstein Telescope
Article References: Schneider, D., & Galler, R. (2026). Excavated Material Strategy for Large Underground Research Infrastructures in Early Planning Phases: A Risk and Quality Management Framework Applied to the Einstein Telescope. EPJ Research Infrastructures, 10(1), Article 20. https://doi.org/10.1007/s41781-026-00172-9
Image Credits: AI Generated
DOI: 10.1007/s41781-026-00172-9
Keywords: Einstein Telescope, gravitational waves, excavated material, tunnelling, circular economy, risk management, quality management, geology, research infrastructure, recycled aggregates, sensor-based characterisation, Euregio Meuse-Rhine
News Source: Grant Pearson. (October 29, 2026). Einstein Telescope Excavated Material Framework Targets Millions of Cubic Metres of Rock. Scienmag.
