1. Project Background

The Yan’an region, characterized by mountainous loess terrain, faces significant slope stability challenges due to natural and anthropogenic factors. The project area is dominated by thick loess coverage with well-developed gullies and vegetation cover below 40%. The rainy season brings frequent and intense rainstorms, which easily generate surface floods carrying large amounts of sediment, resulting in severe soil erosion. Poorly developed drainage systems on slopes lead to inadequate water runoff, allowing rainwater to impact slope surfaces and cause loess caves, subgrade settlement, and other instability issues, posing major safety risks to slope stability. To enhance early warning capabilities and ensure public safety, a pilot slope stability monitoring project was conducted using advanced radar technology. The primary objective was to achieve high-precision, real-time monitoring of surface displacement and environmental parameters.

2. Field Test Introduction

The field test was carried out on a representative slope in the Yan’an mountains, where loess coverage, gullied terrain, and seasonal rainstorms pose significant challenges to slope stability. The core monitoring system employed was the ClairWav-SD1K imaging radar, supplemented by corner reflectors and conventional sensors. As illustrated in the monitoring layout, instrumentation was distributed across the upper and lower slopes, along the highway, and at critical structural interfaces.

Key monitoring parameters included:

• Slope Deformation Monitoring:
  Surface displacement (primary focus), settlement of key slope points, differential settlement, and deep soil deformation.
• Retaining Wall Monitoring:
  Strain of the soil wall, stress on retaining wall anchors, and tilt monitoring of the retaining wall.
• Earth Pressure and Water Pressure Monitoring:
  Groundwater pressure, water level monitoring, and pore water pressure (using vibrating wire crack piezometers).
• Environmental Monitoring:
  Rainfall (via rain gauges), temperature, and humidity monitoring.

The integration of radar-based remote sensing with in-situ sensors (e.g., vibrating wire piezometers, tiltmeters, and temperature/humidity loggers) enabled high-precision, real-time tracking of surface and subsurface changes. Corner reflectors were strategically placed to enhance radar signal quality. This multi-parameter approach supports early warning of loess cave development, subgrade settlement, and retaining wall instability, thereby improving public safety along the highway corridor.

3. Monitoring Scope

The project integrated multiple monitoring dimensions to provide a comprehensive slope stability assessment:

4. System Principle & Deployment

4.1 Principle of ClairWav-SD1K Radar + Corner Reflectors

The ClairWav-SD1K is a ground-based interferometric synthetic aperture radar (GB-InSAR) system that monitors slope surface movement by measuring the radial displacement of passive corner reflectors relative to the radar unit. The system operates by transmitting a sequence of frequency-modulated continuous wave (FMCW) signals in the Ku-band (typically 17.1–17.6 GHz) toward the slope surface. Passive corner reflectors, constructed from trihedral metallic structures, are strategically installed on the slope to provide stable, high-radar-cross-section (RCS) targets.

During each scanning cycle, the radar receives backscattered signals from the corner reflectors and the surrounding terrain. By applying interferometric processing to successive scans, the system extracts the phase difference (Δφ) of the reflected signals for each resolution cell. The phase change is directly proportional to the line-of-sight (LOS) displacement (d) of the corner reflector, governed by the relationship:

d = (λ · Δφ) / (4π)

where λ is the radar wavelength (approximately 1.7 cm in the Ku-band). This allows for millimeter-level displacement accuracy (typically ±0.1 mm to ±1 mm), depending on signal-to-noise ratio (SNR) and environmental conditions.

To further enhance measurement reliability, the system employs:

• Range-Doppler focusing algorithms to isolate individual corner reflectors from clutter and stationary background features.
• Temporal coherence analysis to distinguish true displacement from atmospheric phase noise (e.g., temperature, humidity fluctuations), which is corrected using reference stable reflectors or meteorological sensors.
• Spatial filtering and persistent scatterer (PS) techniques to improve displacement extraction in low-SNR zones.

Any movement of the corner reflector (e.g., due to slope creep, subsidence, or sliding) alters the round-trip path length between the radar and the target, producing a measurable phase shift. The system outputs real-time, two-dimensional displacement maps (radial direction only) at user-defined intervals (e.g., every 5–10 minutes), enabling early detection of slope instability.

This principle combines the wide-area coverage of radar with the high-precision, long-term stability of passive corner reflectors, making the SD1K particularly suitable for monitoring sparsely vegetated loess slopes where point-based sensors (e.g., inclinometers, GPS) are either insufficient or cost-prohibitive for dense spatial coverage.

4.2 Deployment Strategy

The deployment strategy for the ClairWav-SD1K system was designed to maximize measurement accuracy, spatial coverage, and operational reliability under the specific topographic and environmental conditions of the Yan'an loess slope.

• Radar Installation:
  The radar unit was positioned on a stable platform opposite the monitored slope, ensuring an unobstructed line-of-sight (LOS) to the slope surface. The horizontal distance from the radar to the slope face was set to be greater than the total slope length (from crest to toe), typically at a ratio of 1.2:1 to 1.5:1. This configuration minimizes grazing angle effects and reduces geometric distortion, enabling optimal radial displacement sensitivity across the entire slope. The radar was mounted at a height of approximately 2–3 m above ground level to avoid near-field ground clutter and multipath interference. A solar power system with battery backup was deployed to support continuous, off-grid operation during the rainy season.
• Corner Reflector Installation:
  Passive trihedral corner reflectors (with an opening diameter of 30–50 cm) were installed along critical slope sections identified through preliminary geological surveys. The installation intervals ranged from 10 m to 35 m, depending on local deformation risk and surface accessibility. Reflectors were densely spaced (10–15 m) in zones with historical subsidence, loess cave development, or retaining wall interfaces, and more widely spaced (25–35 m) on stable or less critical areas. Each corner reflector was rigidly anchored to the ground using steel posts embedded into a concrete base to ensure long-term positional stability. The reflectors were oriented with their symmetry axis pointing directly toward the radar to maximize radar cross-section (RCS) and signal-to-noise ratio (SNR). Vegetation within a 1–2 m radius around each reflector was cleared prior to installation to prevent signal occlusion.
• Auxiliary Measures:
  To support data interpretation and atmospheric correction, a small weather station (measuring temperature, humidity, and barometric pressure) was co-located with the radar. Additionally, reference corner reflectors were installed on stable bedrock areas outside the slope boundary to serve as phase stability benchmarks.

This deployment strategy balanced technical precision with field constraints, enabling high-quality, real-time displacement monitoring across the heterogeneous loess slope.

4.3 ClairWav-SD1K Imaging Radar Specifications

4.4 Software Platform Capabilities

The monitoring platform is designed to support comprehensive slope stability management with a range of advanced functions. It is capable of monitoring multiple slopes simultaneously, providing real-time data preview from each individual monitoring point. Historical data storage and playback functionality enables trend analysis and post-event investigation. The platform supports integration of multiple sensor types (e.g., radar, inclinometers, piezometers, rain gauges) for data fusion and cross-correlation analysis. It also automatically generates monitoring reports to streamline documentation and compliance. Furthermore, the system supports configurable warning strategies, allowing users to define threshold-based or intelligent assessment rules for automated alerting. This integrated approach enables efficient, multi-parameter early warning and informed decision-making for slope safety.

4.5 System Equipment List

Accessories

5. Benefits

Large monitoring coverage with low equivalent cost per monitoring point

The SD1K system can simultaneously monitor dozens to hundreds of corner reflectors across several thousand square meters in a single setup. Adding additional monitoring points requires only a low-cost passive reflector, whereas point-based sensors (e.g., GPS or inclinometers) require expensive individual units. This makes dense spatial coverage economically feasible for large-scale loess slopes.

High accuracy (≤0.1 mm) enables early detection of micro-deformations

By exploiting radar phase interferometry, the system achieves sub-millimeter displacement accuracy (≤0.1 mm). This sensitivity allows detection of early-stage surface translation in loess slopes before macroscopic cracks or sudden failure occur, providing critical lead time for warnings and mitigation.

All-weather, long-distance operation (≥500 m)

The Ku-band radar signals penetrate light rain, fog, and dust, enabling continuous 24/7 monitoring even during heavy rainstorms—a common trigger of slope instability in Yan'an. With an operational standoff distance of 500 m or more, the radar can be placed on stable ground safely away from the hazardous area.

Low power consumption and solar compatibility for remote deployment

The system consumes only 15–30 W during continuous operation, allowing it to be powered entirely by a small solar panel and battery bank. This enables autonomous, unattended deployment in remote areas without grid access, such as the gullied loess terrain of Yan'an.

6. Challenges Identified

Despite its advantages, the ClairWav-SD1K system exhibited limitations under specific site conditions:

1. Cannot operate in areas with complex vegetation cover – signal attenuation and multipath effects degrade measurement quality.
2. Not suitable for scenarios with dense or complex vegetation – foliage obstructs the radar line-of-sight to corner reflectors.
3. The system cannot function properly under complex vegetation coverage – leading to data gaps or false displacement readings.

7. User Feedback (Post-Project Survey)

“Given the Yan'an region's challenging loess terrain with sparse vegetation (<40%), well-developed gullies, and concentrated rainstorms, the ClairWav-SD1K imaging radar with corner reflectors demonstrated clear advantages over alternative monitoring techniques.”

Key advantages observed in our project:

• Large Coverage with Cost Efficiency: Unlike differential GPS, which requires high-cost per monitoring point, the SD1K provided continuous wide-area surface displacement monitoring with a low equivalent cost per point—critical for covering extensive loess slopes.
• High-Precision Real-Time Monitoring: The radar delivered exceptional deformation accuracy on bare or sparsely vegetated slopes, enabling early detection of surface translation—the dominant mode of early-stage loess landslides. This capability is not achievable with inclinometers, which cannot detect sliding deformation.
• Ideal for Local Conditions: With natural vegetation cover below 40%, the impact of signal blockage was minimal. Compared to satellite radar imaging (long 15–30 day revisit intervals and >1 m corner reflectors), the SD1K offered flexible, on-demand monitoring with smaller, easier-to-install reflectors.
• Operational Efficiency: Lightweight design and solar power support made field deployment straightforward. Real-time alerts and intuitive data visualization further enhanced early warning capabilities for highway safety.

“In summary, for loess regions with sparse to moderate vegetation, the SD1K system provides an optimal balance of coverage, precision, real-time response, and cost—making it a highly effective solution for slope stability early warning.”

8. Conclusion

The Yan'an Slope Stability Monitoring Pilot Project successfully demonstrated that the ClairWav-SD1K imaging radar with corner reflectors is a highly effective solution for real-time slope deformation monitoring in loess terrain. The system delivered exceptional performance on open, sparsely vegetated slopes (<40% vegetation cover), achieving sub-millimeter accuracy (≤0.1 mm) for early detection of surface translation—the dominant precursor to loess landslides. Its large monitoring coverage provided a low equivalent cost per monitoring point, while all-weather operation and solar-powered, low-energy design enabled reliable, autonomous deployment in remote areas with challenging access. The real-time alerts and intuitive data visualization further enhanced early warning capabilities, contributing to improved public safety along the highway corridor. For future deployments in similar loess environments, the system offers a robust, cost-efficient, and high-precision monitoring solution.