Protecting field stations from heat, floods and vandalism

A monitoring station is only useful on the day it is needed, and that is usually the day the weather is worst. Sensor protection is therefore not an accessory; it decides whether the record exists. This article explains how to think about protecting field monitoring stations from the three hazards that cause most losses: heat, floods and people. The common thread is that protection starts with where and how the station is built, long before any cover or cage is added.

What are the main threats to a field station?

The main threats are solar heat, water and debris, lightning, animals and human interference. The WMO Manual on Stream Gauging notes that gauge structures "in many cases ... provide the protection necessary against vandalism and natural hazards such as rain, floods, wind and lightning" [1]. In other words, the shelter, the mast and the mounting position are themselves the first line of defence, and a sensor that is excellent on the bench still fails if it is placed badly.

How do you protect a sensor from heat?

Protect it from heat by keeping direct sunlight off the sensor and the electronics, while still letting air pass over them. Two separate problems are involved.

Measurement error. An air-temperature sensor in direct sun reads the sun, not the air. The WMO Guide to Meteorological Instruments and Methods of Observation recommends that a radiation shield give an enclosure whose internal temperature is uniform and equal to that of the outside air [2]. A Japan Meteorological Agency intercomparison of thermometer shields found that, among four naturally ventilated types, solar radiation produced a deviation of up to +3.4 °C from the reference in the worst shield [2]. The choice of shield, not only the sensor, therefore sets the quality of the reading. We cover siting in siting a weather station: height, obstacles and exposure rules.

Equipment survival. Sealed boxes in full sun can get hot inside, because heat from the sun and from the electronics has little way out. Practical measures are a sun-shield or canopy above the enclosure, a light-coloured finish, space around the box for air to move, and keeping heat-producing parts such as charge controllers away from sensitive ones.

The international series of environmental tests, IEC 60068, defines dry-heat tests (Test B) for checking whether equipment can be used, transported or stored at high temperature, with variants for heat-dissipating equipment that stays powered throughout [3]. When you read a test claim, check which test was run and at what temperature, and whether the unit was powered.

How do you protect a station from floods?

You protect it by putting the equipment where flood water, debris and erosion cannot reach, not by making it sturdy enough to survive them. The WMO manual advises that a shelter for a bubble gauge or pressure transducer be installed "at any convenient location above the reach of floodwaters" [1]. A USGS site-hardening checklist asks the same question in another way: are the shelter and its fittings vulnerable to bank erosion or damage from flood-borne debris? [4]

A worked example, which is illustrative:

Highest flood mark known at the site          = 4.2 m above gauge zero
Allowance for a larger flood and debris wave  = 1.0 m
Lowest safe height of the enclosure base      = 4.2 + 1.0 = 5.2 m above gauge zero

In hill catchments the channel is mobile, so ask local residents and district records, not just the last survey. See river monitoring in hill and Himalayan catchments.

Non-contact sensors help here. A radar mounted on a bridge or arm looks down at the water and never touches it, so floating logs and silt cannot strike it. They still need a strong, rigid mount clear of debris, which we describe in how to install radar level and velocity sensors on a bridge.

Lightning needs its own treatment: grounding, surge protection and cable routing. The WMO manual states that built-in protection should be supplemented where it is inadequate, and that even with layers of protection a direct strike will likely destroy the electronics [1]. We look at this in lightning and surge protection for remote field stations.

How do you stop vandalism and theft at a monitoring station?

Reduce what can be reached, hidden or carried away, and make what remains hard to open. Vandalism is not rare. In January 2026 a river gauge in New Plymouth, New Zealand, had its door ripped off and equipment lost, the site's third attack since 2020, and the regional council warned that such damage "could seriously impact the timeliness of alerts to the public about rising river levels" [5]. In April 2026 equipment was stolen from a Hawke's Bay river monitoring site as a cyclone approached, and the system lost contact with the network [6].

Measures that follow from this:

• Make it unattractive. Solar panels, batteries and copper cables are the usual targets. Mount panels high, keep cables inside conduit, and avoid leaving loose items on the site. For a bubble or pressure gauge, the WMO manual describes encasing the connecting tube or cable in metal conduit or burying it, to protect it from the elements, animals and vandalism [1].
• Use a real enclosure, not a sheet-metal box. The USGS checklist lists purpose-built electrical enclosures rather than sheet metal, and fortified enclosures at sites with a history of vandalism [4].
• Keep the site low-profile. The WMO manual notes, for example, that a coil bridged across a channel for electromagnetic velocity measurement is normally used only where its presence is acceptable and "not subject to vandalism" [1].
• Protect the data too. A device that records on board keeps its record if the network goes down, and a locked, authenticated device is harder to tamper with remotely; see security for field IoT devices.

A short checklist before you commission a station

1. Is the enclosure base above the highest flood mark plus an allowance?
2. Is the site away from eroding banks, and clear of debris paths?
3. Is the enclosure shaded, with free airflow around it, and is the temperature sensor in a proper radiation shield?
4. Are cables in conduit, panels out of casual reach and fixings hard to remove?
5. Are grounding and surge protection fitted and tested?
6. Is a post-monsoon inspection scheduled? See the post-monsoon maintenance checklist for river monitoring stations.

Key points

• Protection is mostly siting and mounting: above flood reach, away from erosion, shaded and ventilated.
• A radiation shield matters as much as the sensor for air temperature; one naturally ventilated shield read up to 3.4 °C high in a JMA test [2].
• Check what a heat test actually covers before relying on a temperature claim [3].
• Vandalism and theft hit warning stations at the worst times; reduce what is exposed and use proper enclosures [5][6].
• Keep power and communication resilient too; see keeping stations reporting in floods.

Where ClimMaTech fits

At ClimMaTech we design stations so that the sensor sits clear of the water and the record stays on the device, because the site, the mount and the enclosure are as much a part of the measurement as the sensor. Tarang is our radar AWLR (Automatic Water Level Recorder), and PRAVAAH combines an AWLR with an SVR (Surface Velocity Radar); both are non-contact and have integrated data logging.

References

1. World Meteorological Organization, Manual on Stream Gauging, Volume I: Fieldwork, WMO-No. 1044, 2010 (Chapter 4: Measurement of stage, sections 4.3 and 4.3.3). PDF
2. T. Aoshima, K. Nakashima, H. Kawamura, M. Kumamoto, T. Sakai, S. Kawano, M. Joko, RIC-Tsukuba (Japan) Intercomparison of Thermometer Screens/Shields in 2009–2010, Japan Meteorological Agency, presented at WMO TECO-2010. PDF
3. International Electrotechnical Commission, IEC 60068-2-2:2025, Environmental testing, Part 2-2: Tests, Test B: Dry heat, edition 6.0, 2025. webstore.iec.ch
4. U.S. Geological Survey, Water Resources Division, Gaging station hardening checklist, CPAC action item, 2000. water.usgs.gov
5. Taranaki Regional Council, Vandalism of monitoring stations putting lives at risk, 22 January 2026 (as published by Scoop). scoop.co.nz
6. 1News, Hawke's Bay river monitoring equipment stolen, 17 April 2026. 1news.co.nz