Automatic Water Sampling: Methods, Trigger Modes and Best Practice

Online sensors tell you what water quality is doing moment by moment. Automatic samplers tell you what was actually in the water at a specific moment — physical evidence that can be analysed in a laboratory and, if necessary, relied on in enforcement. This guide covers how automatic sampling works, how to choose a trigger mode, and the practices that keep samples defensible.

Why Automatic Sampling Still Matters

It is reasonable to ask why sampling is needed when continuous sensors exist. Three reasons:

• Not everything can be measured online. Many regulated substances — heavy metals, pesticides, specific organic compounds — have no practical continuous sensor.
• Laboratory analysis is the reference. Online readings are validated against laboratory results, and those results come from samples.
• Enforcement needs evidence. A sensor trace shows a trend; a preserved, documented sample can support formal action.

Grab versus Composite Samples

Sample type	What it is	What it tells you	Typical use
Grab	A single sample collected at one instant	Conditions at that exact moment	Parameters that change on storage, spot checks
Time-proportional composite	Equal aliquots collected at equal time intervals	Average conditions over the period	Where flow is steady
Flow-proportional composite	Aliquots scaled to flow volume at collection time	Flow-weighted average concentration	Where flow varies — most discharge monitoring
Event-triggered	Collection initiated by a condition such as an exceedance	Conditions during the event	Pollution incident investigation

Why Flow-Proportional Sampling Is Usually Right for Discharges

Consider a discharge that is clean and high-flow for twelve hours, then concentrated and low-flow for one hour. A time-proportional composite would be dominated by the twelve clean hours and would understate the pollution. A flow-proportional composite weights each aliquot by the volume it represents, giving the true flow-weighted average — which is what load calculations and most consent conditions require.

This is why samplers with built-in flow metering, supporting triangular weirs, rectangular weirs, equal-width weirs and Parshall flumes, are standard at industrial discharge points.

Trigger and Sampling Modes

• Timed quantitative: A fixed volume at set intervals — simple and predictable.
• Flow-proportional quantitative: Volume or frequency scaled to measured flow.
• Level proportional: Triggered and scaled by measured water level, useful in sewers and channels.
• Parallel / sequential: Discrete aliquots into separate bottles, for later time-resolved analysis.
• Event or exceedance triggered: Sampling starts when a monitored parameter crosses a threshold — the mode that catches intermittent illegal discharges.

Exceedance-Triggered Sampling Is the High-Value Mode

Many pollution events are short and occur outside working hours. A sampler triggered by a conductivity, pH or turbidity threshold will collect a sample precisely when the anomaly occurs — evidence that would otherwise be lost. Linking a sensor to the sampler’s trigger input is one of the highest-value integrations available in discharge monitoring.

Keeping Samples Representative

Sample Preservation

Between collection and analysis, samples change. Biological activity continues, volatile compounds escape, and metals can adsorb to container walls. Standard controls:

• Refrigeration at around 4 °C — the general default, and the reason portable samplers such as the MT-8000D integrate a settable refrigeration system.
• Chemical preservation where the parameter demands it — acid for metals, specific preservatives for nutrients.
• Documented holding times — every parameter has a maximum interval before analysis, and exceeding it invalidates the result.
• Chain of custody — record who collected, who transported and who received the sample, with times.

Avoiding Cross-Contamination

• Purge the sampling line before collecting — residual water from the previous sample is a real contamination source.
• Use automatic line emptying after each sample, which also prevents algae growth and sediment build-up in the tubing.
• Select tubing material compatible with the target analytes; some compounds sorb to certain plastics.
• Clean and, where appropriate, replace tubing between campaigns.

Siting the Intake

• Place the intake in a well-mixed zone, not a stagnant corner or behind a baffle.
• Keep the intake away from the bed where sediment can be drawn in.
• Respect vertical suction lift limits — typically around 8 m for peristaltic samplers — and horizontal run limits.
• Secure the intake so that flow or debris cannot move it.

Installations in Difficult Locations

Conventional cabinet samplers do not fit everywhere. Manholes, sewers and deep wells are confined, often require confined-space entry procedures, and may have no power.

Purpose-built samplers address this directly. A compact square-column unit such as the MT-8000T is designed for manholes, sewers, ditches and deep wells, running on a DC 16.8 V lithium battery with up to 60 days standby, controlled over 4G so that operators never need to enter the confined space, and protected against power loss by automatically caching the sampling task and resuming when power returns.

Documentation Requirements

A defensible sample is a documented sample. At minimum, record:

1. Date and time of each aliquot collection.
2. Sampling mode and trigger condition used.
3. Volume collected and bottle identification.
4. Flow or level at the time of collection, where flow-proportional.
5. Preservation method and storage temperature.
6. Collector identity and chain of custody.
7. Any deviation from the standard procedure, with the reason.

Samplers that automatically log sampling time, sample volume and equipment status — and synchronise alarm records to the cloud — remove most of this manual burden while improving the record.

Common Mistakes

• Using time-proportional sampling where flow varies widely.
• Exceeding holding times because collection and laboratory schedules are not coordinated.
• No line purging, so the first aliquot contains the previous sample.
• Intake sited in a stagnant zone that does not represent the discharge.
• No documentation of maintenance periods, so data collected during servicing is indistinguishable from valid data.
• Trigger thresholds set so wide that real events never start the sampler.

Conclusion

Automatic sampling complements online monitoring rather than competing with it: sensors provide continuity and trigger the alarm, samplers provide the physical evidence. Choose flow-proportional collection wherever flow varies, use exceedance triggers to catch intermittent events, preserve and document rigorously, and the samples you collect will stand up to scrutiny.

Frequently Asked Questions

How many bottles should a composite contain?

Enough that the composite genuinely represents the period. For a 24-hour discharge with variable flow, collecting aliquots at short intervals into one or more containers is standard; the exact scheme depends on the parameter and the applicable standard.

Why refrigerate samples?

To slow biological and chemical change between collection and analysis. Most parameters have defined holding times that assume refrigeration at around 4 °C; without it, results can change materially before the sample reaches the laboratory.

Can a sampler be triggered by an online sensor?

Yes, and it is one of the most useful configurations available. A pH, conductivity or turbidity threshold can start the sampler automatically, capturing evidence of a short-lived event that would otherwise pass unrecorded.

What limits how far a sampler can draw water?

Vertical suction lift is limited by atmospheric pressure and pump capability — typically around 8 m for peristaltic samplers — while horizontal run is limited by friction losses. Check both against the installation before committing to a design.

Is it worth sampling where continuous sensors already run?

Yes, for two reasons: to validate the sensor readings against laboratory results, and to cover parameters that no online sensor can measure. Periodically comparing the two is what keeps an online dataset trustworthy.