Reclamation Selects Seven Phase 1 Winners in Crack the Case Challenge to Detect Subsurface Cracks

DENVER - The Bureau of Reclamation selected seven Phase 1 winners of the Crack the Case Challenge. The seven phase 1 winners will receive prizes and a chance to further develop prototypes and complete testing on systems to detect subsurface cracks.

The seven teams will advance to Phase 2 to refine and validate their proposed solutions, progressing from a concept to a more developed, testable system. During Phase 2, teams will focus on refining their proposed technical approach, addressing risks identified during Phase 1, and proving their proposed method's feasibility through analysis, modeling, prototyping, or testing. Teams will define system architecture, performance goals, deployment plans, and operational factors for embankment dam environments, considering materials, site constraints, and cultural interference. The Phase 2 goal is to show technical confidence and evidence for Phase 3 demonstrations.

The Phase 1 winners are as follows:

Integrated DAS & EM-SP Crack Detection for Dams, submitted by Mahdi Haddad

This team proposes a novel hybrid sensing framework targeting the vadose zone by integrating active-source electromagnetic-streaming-potential (EM-SP) synergy with passive seismic interferometry via Distributed Acoustic Sensing (DAS). The system moves beyond indirect imaging to directly measure deformation and flow-induced responses within the dam body. This provides a high-fidelity assessment of millimeter-wide cracks at depths up to 50 m.

GAIA - Geophysical Anomaly Inference Algorithm submitted by TAB Technologies

GAIA is a physics-informed, multimodal AI that detects embankment dam cracks by fusing ERT and seismic data. It is trained on a large synthetic dataset generated with DamForge, the team's customizable simulation engine, which was developed to solve the lack of ground truth in real world scenarios. This approach enables robust, scalable crack detection and extends to broader dam monitoring applications beyond crack localization.

Dam Monitoring from Space submitted by InfraSAR team

This team proposes to use Interferometric Synthetic Aperture Radar, or InSAR, to monitor spatial changes in differential settlement of embankment dams that cause soil tension and cracking that is not visible at the ground surface.

Nonlinear Dynamic Crack Signature Tomography submitted by Ziga Gosar

Nonlinear Dynamic Crack Signature Tomography actively excites embankment dams and measures their response using distributed fiber optic sensing (DAS). This team proposes identifying hidden cracks through their unique nonlinear dynamic signatures, including wave distortion and energy loss. By analyzing these responses, the system maps crack location, depth, and connectivity, enabling reliable detection of subsurface defects in clayey soils.

Surface Net Acoustic Computed Tomography (SNA-CT) submitted by HeroGuy's team

A distributed network of active acoustic sensors use pulsed chirps and the resultant subsurface vibration fields to generate high-resolution 3D maps of the materials underground in a few minutes. The resulting dataset is processed to display density variations, material stiffness, water content, and other key material properties. These sensors are implemented as compact, lightweight devices that can be deployed by a single operator informing structural maintenance and operations.

Surface Deformation Laser Array submitted by Team Hydra

The Surface Deformation Array detects subsurface cracks by measuring surface changes. Poles linked by lasers measure both displacement and rotation. A crack changes the sediment water ratio which expresses as a volume change. Three mechanisms aid detection: static strain, softening from fines migration, and swelling as water enters clay along crack paths. ML classifies location and activity. Non-invasive, immune to utility interference, works dry or wet.

CrackSense: Fiber-Optic Early Crack Detection by ESG's Team

This team's solution uses a single fiber-optic cable as a continuous sensor along an embankment dam. DSTS is the primary detection method to measure strain at very fine spacing to flag dry cracks by spotting sharp, localized strain changes along the fiber. DAS complements this by acting as a dense seismic array to map broader changes in stiffness and wave speed over time. We combine the DSTS and DAS indicators into a simple crack-likelihood map to guide follow-up inspection and risk response.

To view the award and learn more about the awardees, visit www.herox.com/CrackTheCase