Who Is Responsible for Cooling-Tower Water Management?

The owner cannot contract away program oversight

CDC describes a water-management program as a structured process for identifying where Legionella could grow or spread and establishing controls, limits, monitoring, corrective actions, verification, and validation. Contractors can perform many tasks, but the owner or responsible authority must ensure the pieces form one functioning program.

Name the program team and authority before dividing work. Facility operations, infection prevention or EHS, mechanical service, water treatment, environmental health, and outside specialists may all have roles depending on the building.

Start with a complete system description and flow map

A water-management team cannot assign effective controls to a system it has not described. Document every tower cell, basin, equalizer, condenser-water loop, pump, bypass, side-stream filter, chemical-feed point, controller, conductivity probe, makeup and blowdown connection, heat exchanger, seasonal branch, drain, overflow, sampling point, and connection to other water systems.

Show normal and abnormal flow paths. Low-use branches, offline cells, cross-connections, temporary hoses, redundant equipment, storage, warm areas, and sections that do not drain can behave differently from the main loop. Mark nearby outdoor-air intakes and publicly accessible areas because drift and aerosol pathways matter during routine operation and cleaning.

Keep the drawing operational rather than decorative. It should help staff locate a control point, isolate equipment, trace an alarm, understand what happens when a pump stops, and decide which vendors must be notified.

• Equipment identifiers that match field labels, controls, work orders, and logs.
• Normal operating, standby, seasonal, bypass, drain, and emergency flow paths.
• Treatment injection, monitoring, alarm, sampling, makeup, and blowdown locations.
• Areas with stagnation, low flow, sediment, warm water, aerosol generation, or difficult access.
• Nearby air intakes, occupied areas, roof access routes, and contractor work zones.
• Named owner of the drawing and a revision process after system changes.

Turn the water-management program into operating instructions

A program is useful only when the team can carry it out during an ordinary shift and during an upset. For each identified hazard, define the control measure, control location, limit or target, monitoring method, frequency, responsible person, required record, immediate corrective action, escalation path, and method for confirming that control was restored.

Verification asks whether the program is being performed as written: Are logs complete? Are instruments calibrated? Were alarms answered? Were corrective actions closed? Validation asks whether the program is effective at controlling the hazard. The qualified team should define both activities rather than treating a stack of vendor tickets as proof of effectiveness.

The program should also state who has authority to stop a tower, order emergency work, communicate with building leadership, retain qualified specialists, and approve return to service. Those decisions are hardest to improvise during an urgent event.

Build a responsibility matrix across vendors

List every routine and nonroutine activity: treatment-system operation, monitoring, calibration, sampling when applicable, basin and fill inspection, fan and pump service, cleaning, disinfection, startup, shutdown, emergency response, alarm handling, recordkeeping, and program review.

Assign a lead, backup, required notification, response time, record location, and acceptance criterion. Avoid shared labels such as by others. Shared work needs a named handoff and confirmation.

Write the boundary of every contractor's scope

A water-treatment provider may monitor chemistry and maintain feed equipment while excluding tower cleaning, mechanical repair, microbiological sampling, emergency disinfection, or after-hours alarms. A mechanical contractor may service fans and pumps while excluding basin cleaning, fill inspection, treatment controls, or water-quality interpretation. The owner should make those boundaries visible before awarding the work.

Require each provider to state routine frequencies, response times, qualifications, subcontractors, laboratory relationships, calibration practices, report timing, emergency availability, documentation format, and exclusions. Define who supplies chemicals, test reagents, sensors, replacement probes, access equipment, lockout support, confined-space controls when applicable, and waste handling.

Where two scopes meet, require a handoff. If the mechanical contractor drains a system, who protects the treatment equipment? If treatment data indicate fouling, who inspects the fill and heat exchangers? If a fan or pump failure changes water conditions, who notifies the program team?

• Name the lead and backup for each task.
• Define normal and emergency response times.
• Identify reports, alarm routing, and the owner-controlled record location.
• State exclusions and work assigned to other parties.
• Require notification when one vendor's work changes another control measure.
• Set acceptance criteria and closure evidence for corrective work.

Mechanical condition and water control affect each other

Scale, corrosion, sediment, biofilm, poor distribution, dead zones, drift, leaks, malfunctioning feed equipment, and temperature affect tower operation and risk control. Mechanical and treatment providers should review conditions together when performance or control results change.

The mechanical scope should include safe access and the basin, fill, distribution, eliminators, fans, drives, pumps, strainers, valves, sensors, controls, and nearby system interfaces. The treatment scope should identify parameters, equipment, chemicals, monitoring, alarms, and corrective actions.

Inspect the tower as a working air-and-water system

A cooling tower distributes warm water across fill while moving air through the unit. Heat and some water leave by evaporation, dissolved minerals concentrate in the remaining water, and drift eliminators limit droplets carried out with the air stream. Mechanical condition, water chemistry, airflow, and heat load therefore interact continuously.

Inspection should address the cold-water basin, distribution nozzles or pans, fill, drift eliminators, louvers, casing, access doors, fan, drive, gearbox or bearings, motor, structural supports, strainers, equalizers, makeup assembly, overflow, blowdown, treatment equipment, sensors, and visible piping. Look for sediment, scale, corrosion, biofilm, leaks, damaged fill, poor distribution, blocked strainers, standing water, unusual vibration, and unsafe access.

The inspection frequency should reflect system design, operating season, risk, manufacturer guidance, program requirements, and observed condition. Photographs should use consistent equipment identifiers and viewpoints so the team can distinguish a new condition from one that has been open for months.

Connect treatment data to equipment and operating conditions

Treatment control is more than chemical delivery. The program should identify how makeup-water quality, evaporation, conductivity or another concentration indicator, blowdown, disinfectant strategy, pH and other program-specific parameters, corrosion control, scale control, and biological control are monitored and acted upon.

Cycles of concentration can improve water efficiency, but only within the limits established for the makeup water, system materials, treatment program, and operating conditions. A failed makeup valve, stuck blowdown valve, empty chemical drum, disabled pump, dirty probe, lost controller signal, or low-load operating change can alter control even when the last service report looked normal.

Trend operating and treatment data together. Relate excursions to heat load, makeup use, blowdown, tower staging, weather, shutdowns, cleaning, leaks, and equipment failures. The goal is to detect why control was lost, not merely document that a number was restored.

Out-of-range results need predetermined action

CDC guidance emphasizes control measures, control limits, and control points. The program should state what result is outside range, who receives the alarm or report, what immediate action occurs, when the system is escalated, who can take it offline, and how return to control is documented.

Do not bury recurring excursions in separate vendor reports. Trend them with operating changes, outages, maintenance, weather, water interruptions, and tower condition.

Design records for decisions, not storage

A useful record connects the reading to the instrument, location, date and time, operating state, person, applicable limit, action, notification, follow-up result, and closure. An isolated number without tower status or corrective-action history is difficult to interpret later.

Define instrument calibration and maintenance. Record controller failures, probe cleaning or replacement, loss of communications, chemical-delivery interruptions, alarm overrides, manual operation, and missed rounds. Automated trends improve visibility but do not replace field observation or review by the qualified team.

The owner should control access to the complete record even when vendors maintain separate portals. Establish retention, export, backup, and vendor-transition procedures so history does not disappear when a contract ends.

• Operating status and active tower cells at the time of the reading.
• Parameter, location, method, instrument, and calibration status.
• Applicable limit or target and whether the result was in range.
• Immediate action, responsible person, notification, and response time.
• Follow-up evidence showing return to control and formal closure.
• Links to related work orders, photos, laboratory reports, and vendor visits.

Cleaning and emergency disinfection require a written procedure

CDC's cooling-tower module describes considerations for offline emergency cleaning and disinfection, including reviewing the treatment program, removing heat load, shutting fans, managing feed and blowdown, maintaining makeup water, and closing nearby air intakes during the procedure. Site conditions and qualified expertise still govern the actual method.

Define triggers, authority, isolation, worker protection, chemicals, environmental and waste requirements, communication, documentation, and return-to-service criteria before an urgent event.

Control seasonal startup, shutdown, and disruptions

Startup after seasonal layup, construction, water interruption, treatment failure, extended low load, or an unplanned outage is not routine operation. The program should define inspection, cleaning, treatment, circulation, monitoring, notification, and return-to-service requirements for each relevant condition.

During planned shutdown, decide whether the system will be drained, kept dry, maintained wet under a controlled program, or managed another way under qualified guidance. Avoid leaving warm, stagnant water in equipment that staff assume is safely offline.

Create triggers for unplanned events: loss of makeup, failure of blowdown or feed, tower-cell isolation, pump outage, controls failure, power loss, building closure, extreme weather, emergency cooling, construction debris, or an extended period outside control limits. The procedure should say who assesses the system before restart.

Place testing and sampling inside the program—not beside it

Testing can support verification and validation, but no single laboratory result replaces system control. The program team should define why a sample is collected, where and how it is collected, which method and laboratory are used, how results are interpreted, who is notified, and what actions follow each result.

Sampling plans should account for the question being asked and the limitations of the method. Avoid creating an automatic pass-or-fail rule from a result that the qualified team has not connected to system conditions, control history, applicable guidance, and response procedures.

Preserve chain-of-custody and contextual information when laboratory testing is used. Record tower status, treatment conditions, recent cleaning or disinfection, sampling point, collection method, transport, and related operational events.

Procure one outcome across multiple contracts

Issue the same system description, program requirements, control framework, reporting format, and escalation contacts to every provider. Require vendors to identify exclusions, subcontractors, laboratory relationships, assumptions, and information they need from others.

During proposal review, walk through a missed alarm, failed chemical feed, fan outage, dirty basin, water interruption, extended shutdown, and suspected case response. The exercise exposes gaps that a price table will not.

Use a short owner audit to keep the program alive

At a defined interval, the program team should confirm that the system map is current, responsible people are still available, vendor scopes still close every gap, limits and responses remain appropriate, monitoring is complete, instruments are maintained, and corrective actions are closed.

Review changes in equipment, treatment, water supply, operating schedule, occupancy, building use, construction, staffing, vendors, remote monitoring, and applicable requirements. A program copied forward without these checks can look complete while describing a building that no longer exists.

• Can staff locate the current program, drawings, contacts, and emergency procedures?
• Do recent logs show timely response and documented closure when limits were missed?
• Are mechanical findings visible to the treatment provider and program team?
• Are calibrations, alarm tests, inspections, cleaning, and training current?
• Can the owner export and retain all vendor data and reports?
• Have system, staffing, or operational changes been incorporated into the hazard analysis?

The bottom line

Cooling-tower risk increases in the spaces between scopes. A strong program gives the owner one view of water control, mechanical condition, alarms, corrective actions, and current responsibility.

Keep the program active through staff changes, vendor changes, seasonal transitions, construction, outages, and equipment replacement. Review it when the system or operating conditions change, not only after a problem.