MONITOR. DETECT. UNDERSTAND. RESPOND.
Geohazard Monitoring in Hong Kong
Geohazard monitoring in Hong Kong for slopes, landslides and ground movement, integrating instrumentation, groundwater monitoring, remote sensing and engineering intelligence by GEOOE.
Hong Kong Context
Why Geohazard Monitoring Matters in Hong Kong
Hong Kong combines steep natural terrain, a large inventory of engineered slopes, dense urban development and critical transport infrastructure with intense seasonal rainfall. The result is a monitoring problem in which rainfall, groundwater and ground movement often need to be interpreted together rather than as isolated measurements.
The Geotechnical Engineering Office (GEO) of Hong Kong’s Civil Engineering and Development Department (CEDD) manages landslide risk through slope safety programmes, landslide investigation, mitigation and warning systems. Hong Kong’s Landslip Warning System uses a territory-wide network of automatic raingauges and rainfall–landslide relationships to assess changing landslide risk.
GEOOE approaches a Hong Kong geohazard monitoring programme from the failure mechanism and the engineering decision first. The question is not simply which sensor can be installed, but what movement, groundwater response or trigger condition must be measured—and what action that information is intended to support.
Hazard Types
Geohazards and Ground Risks That May Require Monitoring
Slope & Landslide
Movement of natural or engineered slopes where the monitoring objective may include deformation rate, shear-zone development and groundwater response.
Ground Movement & Subsidence
Vertical or lateral ground deformation affecting buildings, roads, utilities, retaining structures or sensitive infrastructure.
Rockfall & Debris Movement
Rock faces and natural terrain where spatial mapping, movement detection and rainfall context can support risk management.
Groundwater-Driven Instability
Changes in groundwater level or pore-water pressure that can influence slope behaviour, particularly during prolonged or intense rainfall.
Post-Earthquake Deformation
Targeted monitoring of slopes, retaining structures and sensitive assets when seismic shaking may have changed the existing ground condition.
Natural Terrain & Infrastructure Exposure
Terrain hazards affecting roads, railway corridors, utilities, buildings and other assets located below or adjacent to steep ground.
Monitoring Strategy
What Should a Geohazard Monitoring System Actually Measure?
A stronger monitoring design starts with the engineering question rather than a predefined instrument list.
Is movement occurring?
Establish whether movement is present, its magnitude, direction and rate of change.
Where is movement occurring?
Distinguish surface displacement from deeper subsurface deformation and possible shear zones.
What is driving the change?
Correlate movement with rainfall, groundwater or other environmental and construction-related factors.
What decision should follow?
Define inspection, review and project-specific action requirements before the monitoring system is deployed.
Instrumentation
Typical Instruments for Geohazard Monitoring
GEOOE may combine manual, automated and remote-sensing methods depending on the failure mechanism, required response time, site access and monitoring duration.
| Parameter | Typical Method | Best Suited For | Key Advantage | Main Limitation |
|---|---|---|---|---|
| Subsurface lateral movement | Manual borehole inclinometer | Periodic deformation profiling | Identifies movement with depth | Requires site access and manual reading |
| Continuous subsurface deformation | In-place inclinometer | Higher-frequency automated monitoring | Remote time-series data | Greater system and maintenance complexity |
| Pore-water pressure | Vibrating-wire piezometer | Local pressure response | Suitable for automated logging | Represents conditions at instrument location |
| Groundwater level | Standpipe piezometer | Long-term groundwater regime | Simple and robust | Response may be slower than pressure sensors |
| Relative displacement | Extensometer / crackmeter | Cracks or defined movement zones | Direct local displacement measurement | Does not represent whole-slope movement |
| Surface displacement | Survey prism + total station | Multiple visible monitoring points | Three-dimensional survey information | Requires line of sight |
| Surface displacement | GNSS | Long-term three-dimensional movement | Independent of total-station sight lines | Requires suitable satellite visibility |
| Surface geometry | LiDAR / photogrammetry / UAV survey | Rock faces and larger terrain areas | Spatial coverage | Survey interval and visibility constraints |
| Regional surface deformation | InSAR | Wide-area deformation screening | Large spatial coverage | Not a direct substitute for subsurface instrumentation |
| Rainfall | Automatic rain gauge | Rainfall-trigger correlation | High-frequency trigger context | Does not directly measure slope response |
Method Selection
Same Parameter, Different Instruments: Which Method Fits the Risk?
Ground and slope movement: inclinometer vs GNSS vs survey vs remote sensing
A manual borehole inclinometer is particularly useful when the engineering question concerns deformation with depth and the possible location of a subsurface shear zone. An in-place inclinometer addresses a similar type of subsurface behaviour but supports higher-frequency automated monitoring.
GNSS and survey prisms instead measure surface movement. GNSS is useful where satellite visibility is adequate, while automated total-station monitoring can efficiently observe many targets where stable instrument positions and line of sight are available.
LiDAR, photogrammetry and InSAR provide much broader spatial coverage. Their value is different: they can reveal distributed surface change or support screening, but they do not automatically identify a deep shear surface. GEOOE therefore treats remote sensing and in-situ instrumentation as complementary tools rather than interchangeable technologies.
Groundwater: standpipe vs vibrating-wire piezometer vs water-level logger
A standpipe is primarily used to observe groundwater level within the response characteristics of the installation. A vibrating-wire piezometer measures pore-water pressure at a defined sensor location and is readily integrated with automated dataloggers.
A water-level logger can automate measurements in a standpipe or observation well. These approaches are not universally interchangeable: groundwater level, local pore pressure and unsaturated-zone response answer different hydrogeological questions.
Tilt and deformation: tiltmeter vs inclinometer vs survey
A tiltmeter records angular change at a local structure or surface. A borehole inclinometer produces a deformation profile with depth, while survey and GNSS systems observe movement of defined surface points. The correct choice depends on whether the required information is rotation, subsurface deformation or global displacement.
Cracks and local movement: crack gauge vs crackmeter vs extensometer
Manual crack gauges are suitable for simple periodic observations. Digital crackmeters and extensometers support higher-frequency measurement of local relative displacement. These measurements should not be confused with whole-slope or regional ground movement: a local crack can change while the broader terrain behaves differently.
Monitoring Architecture
A Practical Monitoring Architecture for Hong Kong Geohazards
Not every site requires every sensing layer. GEOOE selects the architecture according to hazard mechanism, consequences, response time, access and maintenance needs.
Trigger Conditions
Rainfall, weather and other site-specific triggering conditions.
Hydrogeological Response
Groundwater level, pore-water pressure and, where relevant, unsaturated-zone response.
Subsurface Deformation
Inclinometers, in-place sensors and extensometers where movement with depth is important.
Surface Movement
GNSS, survey, tilt, crack, LiDAR, photogrammetry or satellite-based observations.
Engineering Review
QA/QC, trend assessment, trigger review, inspection and project-specific engineering action.
Data Interpretation
Monitoring Is Useful Only When It Leads to a Decision
A geohazard monitoring system should distinguish baseline behaviour from meaningful change. This requires sensor QA/QC, comparison with previous measurements, rate-of-change assessment and, where appropriate, correlation between rainfall, groundwater and deformation.
GEOOE does not recommend generic numerical trigger levels for every slope. Trigger values and action plans should be project-specific and established or reviewed by the responsible geotechnical professionals according to the ground conditions, failure mechanism, design assumptions, baseline behaviour and consequences of failure.
Expected Behaviour
Data remain within the expected baseline range and monitoring continues under the project monitoring plan.
Review Required
A change in trend, unusual response or data anomaly requires validation, comparison with other sensors and potentially increased observation.
Engineering Response
Project-specific trigger criteria may require inspection, engineering review, mitigation or other actions defined in the monitoring plan.
Hong Kong Lessons
Real Geohazard Events That Shape Monitoring Strategy
These are public-sector reference cases. They are not presented as GEOOE projects.
Yiu Hing Road, Shau Kei Wan
CEDD’s detailed study records two landslides on the natural hillside above Yiu Hing Road during the record-breaking 8 September 2023 rainstorm. The combined estimated failure volume was about 4,000 m³, including a large structurally controlled rockslide close to high-rise residential buildings.
Engineering lesson: Hong Kong geohazard monitoring must consider natural-terrain hazards as well as engineered slopes, particularly where transport corridors and densely occupied urban assets lie downslope.
Source: CEDD GEO Report No. 377 ↗
Shek O Road
CEDD’s detailed investigation records landslides on 8 and 14 September 2023 at a roadside fill slope at Shek O Road. The first failure caused part of the northbound lane to collapse, while a second failure occurred during another major rainstorm several days later.
Engineering lesson: road corridors can remain exposed after an initial failure. Post-event monitoring and reassessment may therefore be as important as pre-failure monitoring.
Source: CEDD GEO Report No. 376 ↗
International Reference Cases
Verified Geohazard Monitoring Case Studies
The following projects are referenced from public, first-party or government sources. They are not presented as GEOOE projects.
Cleveland Corral Landslide, California
USGS conducted near-real-time monitoring of the episodically active Cleveland Corral landslide near U.S. Highway 50. Published monitoring data include piezometers for pore-water pressure, extensometers for downslope displacement, volumetric water-content sensors and a rain gauge.
Lesson: hydrological response and displacement can be interpreted together rather than as separate data streams.
Source: U.S. Geological Survey ↗
Yui Landslide Monitoring
Japan’s Ministry of Land, Infrastructure, Transport and Tourism describes monitoring in the Yui area using devices for both surface and underground movement, together with precipitation and groundwater observations. Some instruments operate through round-the-clock remote monitoring, and a borehole tiltmeter is used for periodic field observation.
Lesson: surface movement, subsurface behaviour and hydro-meteorological conditions provide different but complementary information.
Source: MLIT Mt. Fuji Sabo Office ↗
Takisaka Landslide, Fukushima
A Public Works Research Institute publication describes three ground-surface movement monitoring systems installed at the Takisaka landslide: optical-fibre sensing, Hybrid GPS and a three-dimensional laser scanner.
Lesson: point-based displacement and spatial surface measurement can be combined when a single sensing technique cannot answer every monitoring question.
Tanjong Rimau Slope Stabilisation
Sentosa Development Corporation’s Tanjong Rimau study addresses deteriorating coastal slope conditions, erosion and rockfall. Proposed stabilisation measures include soil nails, grid beams, rock dowels, rock netting and toe protection.
The published environmental monitoring and management plan covers biodiversity, water quality, air quality, noise and ground vibration during and after the works.
Lesson: geohazard projects can require environmental and geotechnical monitoring to operate as one coordinated project-control framework.
GEOOE deliberately does not attribute unverified instruments or monitoring systems to projects where the public source does not document them. This distinction is important when using international projects as engineering references.
GEOOE Approach
How GEOOE Approaches Geohazard Monitoring
Decision Before Instrument
GEOOE starts with the hazard mechanism, engineering question and required response rather than forcing every project into the same sensor configuration.
Manual + Automated
Periodic manual observations can coexist with automated instrumentation where higher-frequency information is justified by project risk and operational requirements.
Multi-Sensor Interpretation
Movement, groundwater and rainfall should be interpreted together where the failure mechanism requires it, instead of treating every sensor as an independent alarm channel.
Open Monitoring Architecture
GEOOE’s technology direction includes distributed, retrofit-friendly monitoring architecture and data-access concepts that can complement established field instrumentation.
Engineering Review
Data quality, trend interpretation and project- specific trigger review remain engineering functions. Automation should support those decisions rather than obscure them.
Hong Kong Context
Through GEOORIGIN ENGINEERING LIMITED, GEOOE can support project-specific discussion of monitoring strategy, instrumentation planning, data review and engineering coordination in Hong Kong.
Frequently Asked Questions
Geohazard Monitoring FAQ
What is geohazard monitoring?
Geohazard monitoring is the systematic measurement of ground, slope, groundwater and environmental behaviour where natural or ground-related processes could affect people, buildings or infrastructure. The monitoring system should be designed around a defined hazard mechanism and engineering decision.
What instruments are commonly used for slope monitoring in Hong Kong?
Depending on the site, monitoring may include inclinometers, piezometers, standpipes, extensometers, crackmeters, survey prisms, GNSS, tiltmeters, automatic rain gauges and remote-sensing techniques. Not every project requires all of these methods.
What is the difference between an inclinometer and GNSS?
A borehole inclinometer measures lateral deformation with depth and can help identify a subsurface deformation zone. GNSS measures movement of a surface point in three dimensions. They therefore answer different engineering questions and can be used together.
What is the difference between a standpipe and a vibrating-wire piezometer?
A standpipe is typically used to observe groundwater level, while a vibrating-wire piezometer measures pore-water pressure at a particular sensor location. Their response characteristics and interpretation differ, so selection should follow the hydrogeological question.
Can InSAR replace ground instrumentation?
Not in general. InSAR can provide valuable wide-area surface-deformation information, but it does not directly measure subsurface shear-zone development or local pore-water pressure. GEOOE considers satellite observations complementary to in-situ instrumentation where both are justified.
When should automated slope monitoring be used?
Automation becomes particularly useful where the required monitoring frequency, access constraints, consequence of failure or need for rapid engineering review makes periodic manual measurements insufficient. Automation should still include appropriate QA/QC and maintenance provisions.
How does rainfall affect landslide monitoring in Hong Kong?
Intense and prolonged rainfall can alter groundwater and pore-water pressure, which may influence slope behaviour. For this reason, rainfall data are often interpreted together with groundwater and movement measurements rather than viewed as an isolated indicator.
How are monitoring trigger levels established?
Trigger levels should be project-specific. They should reflect design assumptions, baseline behaviour, expected movement, failure mechanism, instrument performance and the consequences of the observed change. Universal trigger values are not appropriate for all slopes or geohazard projects.
Technical References
Sources & Case Study References
Hong Kong
CEDD — Geotechnical Engineering Office ↗
Hong Kong Slope Safety — Landslip Warning System ↗
International
USGS — Cleveland Corral Landslide ↗
Japan MLIT — Yui Landslide Monitoring ↗
Technical guidance on this page is intended for application discussion. Monitoring design, instrument selection and trigger levels should be established for the specific site conditions, failure mechanisms and project requirements by appropriately qualified engineering professionals.
GEOOE · HONG KONG
Planning a Geohazard Monitoring Programme?
Every slope, landslide and ground-movement problem has a different mechanism, consequence and monitoring objective. GEOOE and GEOORIGIN ENGINEERING LIMITED welcome discussions with asset owners, consultants and contractors on practical monitoring strategies for Hong Kong projects.