Water level and flow monitoring at the Nyingchi Mountain Flood Hydrological Station in Xizang

I. Project Background

Mountain flood disasters are characterized by strong suddenness, great destructive power and high difficulty in prediction and early warning. They are one of the most serious natural disasters in mountainous areas of China. According to data from the Ministry of Emergency Management, in 2024, the direct economic losses caused by mountain floods in China will exceed 30 billion yuan, and the number of deaths and dispersons due to disasters will account for 65% of the total casualties caused by floods. The terrain in mountainous areas is complex and the climate is changeable. Traditional monitoring methods are difficult to cover remote areas, and the transmission of early warning information lags behind, which has become a prominent shortcoming in mountain flood prevention. The state attaches great importance to the prevention and control of mountain flood disasters. The "14th Five-Year Plan for the National Emergency Response System" clearly states that "the monitoring and early warning network for mountain flood disasters should be improved to achieve full coverage of monitoring in key hidden danger areas", and the "Technical Specifications for Mountain Flood Disaster Prevention and Control Engineering" requires "the construction of automated monitoring stations in areas prone to mountain floods to enhance real-time monitoring capabilities". Against this backdrop, the construction of mountain flood field monitoring station equipment, which captures key data such as rainfall, water level and soil moisture through intelligent means, and builds a rapid response chain of "monitoring - early warning - evacuation", is of great significance for reducing the losses caused by mountain floods and safeguarding the lives and property of the people.

Ii. Current Problems

Insufficient monitoring coverage: 70% of the areas prone to mountain floods are located in remote mountainous regions. The current density of monitoring stations is less than one per 100 square kilometers, and some areas are completely blind spots in monitoring. In 2023, a mountain flood disaster in a certain county in Southwest China was delayed by 2 hours due to the lack of monitoring data, resulting in 12 deaths.

Poor adaptability of equipment: Field monitoring equipment is affected by extreme environments such as heavy rain, lightning, and high temperatures, with a failure rate as high as 40%. In northern mountainous areas, the low temperature in winter shortens battery life to 3 days, and data transmission interruptions occur frequently.

Insufficient data accuracy: Traditional rain gauges have a measurement error of over 20% during heavy rainfall (> 50mm/h). Water level monitoring relies on manual scales, which cannot capture the process of sudden flood rise, resulting in an unreasonable setting of warning thresholds.

Weak power supply and communication: 60% of the field stations have no municipal power supply, and the solar power supply system often shuts down due to insufficient power during continuous rainy weather. The 4G signal coverage rate in mountainous areas is only 55%, and the data transmission interruption rate exceeds 30%.

Lagging maintenance response: After equipment failure, manual inspection takes an average of 2 days, and for remote stations, it can even take more than 5 days. In 2024, a monitoring station in a mountainous area of South China missed three opportunities for mountain flood warnings due to sensor failure.

Iii. Solution Approach

Optimize the layout of monitoring stations: In line with the mountain flood risk zoning, increase the frequency of monitoring stations at the exits of gullies, the confluence of rivers, and 1-3 kilometers upstream of villages. The monitoring density in key areas should be raised to 1 station per 50 square kilometers to achieve full coverage of potential hazard points.

Strengthen the environmental adaptability of equipment: Industrial-grade weather-resistant equipment is selected, with a protection level of ≥IP66, and the operating temperature range is expanded to -40 ℃ to 70℃. It is equipped with a lightning protection grounding system (grounding resistance ≤4Ω) to reduce the failure rate in extreme environments.

Enhance the accuracy of monitoring data: By adopting high-precision sensors, the rainfall monitoring error is ≤5% (during heavy rainfall), the water level monitoring accuracy reaches ±1cm, and the soil moisture data collection interval is shortened to 10 minutes, capturing subtle changes.

Build a diversified guarantee system: Adopt a hybrid power supply of "solar energy + lithium battery + backup generator", with continuous rainy days of ≥15 days. In terms of communication, dual-mode transmission of "4G/5G + Beidou short message" is achieved to ensure that data is not interrupted. Establish an intelligent operation and maintenance mechanism: Through the remote monitoring system for equipment status, issues such as low battery voltage and sensor failure are warned in advance. Equipped with a drone inspection team, the fault response time is shortened to within 24 hours.

IV. Hardware Perception System

1.Meteorological monitoring equipment

Tipping bucket rain gauge: Core monitoring of rainfall intensity and cumulative rainfall, resolution 0.5mm, supports heavy rainfall measurement at 100mm/h, built-in heating device to prevent freezing in winter, data output frequency 1 minute per time.

Meteorological station: Monitors wind speed (0-60m/s), wind direction (0-360°), air temperature (-40℃-60℃), and humidity (0-100% RH), providing auxiliary parameters for rainfall flow generation models, with a protection level of IP65.

2. Hydrological monitoring equipment

Radar water level gauge: Non-contact measurement of water levels in rivers and gullies, with a range of 0-50m, accuracy of ±3mm, resistant to interference from sediment and floating objects, suitable for turbid water flow environments, and fast data update frequency.

Ultrasonic flowmeter: Monitors flow changes in key river sections, with a measurement range of 0.1-10m/s, supports non-full pipe/full pipe measurement, and has an internal data storage function.

3. Soil and geological monitoring equipment

Soil moisture sensor: Buried at depths of 50cm and 100cm, it monitors the soil volumetric water content (0-100%) and temperature (-30℃-50℃), determines the soil saturation degree, and predicts the slope convergence rate.

Landslide crack gauge: Installed on landslide bodies, it measures the opening degree of cracks (range 0-50mm, accuracy ±0.1mm), perceives the deformation trend of geological bodies, and provides early warnings of landslide risks.

4. Data transmission and power supply equipment

Data acquisition terminal: Integrates multiple sensor interfaces, supports RS485 and analog signal access, has data preprocessing and encryption functions, and has a large local storage capacity.

Communication module: The 4G/5G module supports full network compatibility. The Beidou short message module has a single communication volume of ≥100 bytes and automatically switches when there is no ground network to ensure the priority transmission of early warning information.

Power supply system: 60W monocrystalline silicon solar panel (conversion efficiency ≥23%) +12V/100Ah lithium iron phosphate battery (cycle life ≥2000 times), equipped with MPPT controller and lightning protection module. Battery heating function is enabled in low-temperature environments.

5. Auxiliary equipment

Solar power supply box: Waterproof grade IP66, built-in charge and discharge management system, supports remote viewing of battery status, and has overcharge, overdischarge and short circuit protection.

Anti-theft bracket: Made of 304 stainless steel, it is equipped with an anti-tampering alarm sensor. When the device is moved or damaged, it will automatically send an alarm message to the management platform.

The array radar wave flow measurement system assists hydrological stations in mountain flood monitoring and early warning

V. Scene Application

Mountain flood warning response: A monitoring station in a certain mountainous area detected a rainfall of 80mm within one hour. The soil moisture sensor indicated that the surface soil moisture content was saturated (> 90%), and the radar water level gauge data soared from 1.2m to 3.5m within five minutes. The system automatically triggered a red alert and sent the alert information to the emergency office of the township via Beidou short message. At the same time, it was linked with the village broadcasting system and the Tongluo warning system. Within 30 minutes, more than 500 villagers were evacuated, avoiding casualties.

Hydrological station flow measurement system

Extreme weather monitoring: During the heavy snowfall in northern winter, the monitoring station activates the heating function to prevent the rain gauge from freezing and operates continuously in an environment of -25 ℃, accurately recording the snowfall volume (equivalent to a rainfall of 12mm). Based on the data of wind speed (8m/s) and soil freezing depth, predict the risk of ice floes blocking rivers during the snowmelt period and notify relevant departments in advance to clear the rivers.

Landslide linkage early warning: The landslide crack meter detected that a certain mountain's crack expanded by 3mm in a single day, and at the same time, the soil moisture content indicated that the deep water content was continuously rising. The system judged the landslide risk level as "high". The data was pushed to the geological disaster monitoring center. Combined with the rainfall forecast (heavy rain is expected in the next 24 hours), the temporary evacuation of 30 households around the landslide site was initiated. Two days later, a small-scale landslide occurred in the area, but there were no casualties.

Intelligent operation and maintenance of equipment: The management platform detected through remote monitoring that the charging efficiency of solar panels at a certain site had dropped to 70%, and judged that there might be dust coverage or Angle deviation. The drone inspection confirmed the dust accumulation on the photovoltaic panels. The operation and maintenance team carried cleaning tools to handle it. The equipment was restored to normal power supply within two hours to avoid monitoring interruption due to insufficient power.

Data-driven decision-making: Through the accumulation of three years of monitoring data, the water conservancy bureau of a certain county analyzed and concluded that when the hourly rainfall intensity is ≥40mm and the soil moisture content is ≥80%, it is very likely to trigger mountain floods. Based on this, the early warning threshold was adjusted, raising the early warning accuracy rate from 65% to 88% and extending the early warning advance time from 15 minutes to 25 minutes, thus providing more sufficient time for the evacuation of the public.

This project plan aims to achieve "early detection, early warning and early handling" of mountain flood disasters by establishing a mountain flood field monitoring system that is "highly adaptable to the environment, highly accurate in data and stable and reliable in transmission".