Instrumentation, Operation and Dam Safety Management

Introduction and Context

Dams are long-life, high-consequence infrastructure systems. Their safety depends not only on sound design and construction but also on continuing observation, competent operation, timely maintenance, and effective emergency management. Instrumentation provides evidence about the behaviour of the dam, foundation, abutments, reservoir, spillway and appurtenant works. Surveillance converts that evidence into engineering judgement; dam safety management converts judgement into decisions, actions and accountability.

Modern dam safety is therefore a socio-technical activity. It combines geotechnical, structural, hydraulic, hydrological and environmental engineering with data management, risk analysis, organisational governance and emergency preparedness. No instrument can prove that a dam is safe, and no single abnormal reading should automatically be interpreted as failure. Reliable decisions require knowledge of baseline behaviour, trends, uncertainties, loading conditions and credible failure modes.

Learning Objectives

After studying this chapter, the reader should be able to:

1. Explain the purpose, selection and limitations of dam instrumentation.

2. Develop monitoring and surveillance systems for seepage, pore pressure, deformation, stress, vibration and environmental loading.

3. Interpret piezometric, deformation and seepage records using baseline behaviour, trends and thresholds.

4. Apply inspection, failure-mode analysis and risk-informed prioritisation.

5. Describe operational controls, emergency action planning and incident response.

6. Evaluate rehabilitation, data governance and emerging digital technologies in dam safety management.

1. Dam Safety as a Lifecycle Management System

A dam safety programme should operate through the entire asset lifecycle:

• Design and construction: identify hazards, establish design assumptions, install instruments and record as-built conditions.
• First filling and commissioning: observe the response of the dam and foundation under gradually increasing hydraulic loading.
• Routine operation: collect, validate and interpret data; inspect visible conditions; maintain equipment and records.
• Periodic review: reassess hydrology, seismic hazard, material condition, operating rules and failure consequences.
• Modification or decommissioning: control risks introduced by alterations, ageing or changes in downstream development.

Safety management is based on the interaction of three elements:

1. Loading: reservoir level, seepage gradients, temperature, rainfall, earthquakes, floods, construction activity and operational transients.

2. Response: pore pressure, seepage, displacement, stress, cracking, vibration and visible deterioration.

3. Decision: continued operation, increased observation, engineering investigation, drawdown, repair or emergency action.

A monitoring programme is effective only when each important measurement is linked to an operational decision. Collecting data without defined interpretation procedures creates the appearance of control without necessarily improving safety.

2. Instrumentation Systems

2.1 Purposes of instrumentation

Instrumentation is used to:

• verify design assumptions;
• establish baseline or reference behaviour;
• detect changes before they become visually obvious;
• assess the effectiveness of drainage, grouting and remedial works;
• distinguish normal seasonal response from abnormal behaviour;
• support safe operation during floods, earthquakes and rapid reservoir changes; and
• provide evidence for risk assessment and regulatory review.

Instrumentation should be selected from identified failure modes rather than installed merely because a particular device is customary.

2.2 Main instrument groups

Piezometric and pore-pressure instruments include standpipe piezometers, vibrating-wire piezometers, pneumatic piezometers and observation wells. They measure hydraulic head or pore-water pressure in foundations, embankments, abutments and drainage zones. Vibrating-wire instruments are suitable for automated systems but require attention to temperature effects, cable integrity and calibration.

If a piezometer measures pore-water pressure \(u\), the equivalent pressure head is

\[

h_p=\frac{u}{\gamma_w}

\]

where \(h_p\) is pressure head, \(u\) is pore pressure in pascals, and \(\gamma_w\) is the unit weight of water in newtons per cubic metre. If elevation datum is \(z\), total hydraulic head is

\[

H=z+\frac{u}{\gamma_w}

\]

This relationship assumes consistent datum and units and neglects velocity head, which is appropriate for slowly moving groundwater in dam foundations.

Seepage instruments include weirs, calibrated measuring sections, V-notch boxes, drainage-flow meters, turbidity sensors and leakage collection systems. For a sharp-crested, fully contracted 90-degree V-notch weir, a commonly used relationship is

\[

Q=C_d\frac{8}{15}\sqrt{2g}\tan\left(\frac{\theta}{2}\right)H^{5/2}

\]

where \(Q\) is discharge, \(C_d\) is the discharge coefficient, \(g\) is gravitational acceleration, \(\theta\) is notch angle and \(H\) is head above the vertex. The equation applies only when the weir geometry, approach conditions, aeration and submergence satisfy the assumptions of the calibration. Field calibration and sediment control remain essential. Seepage quantity must be interpreted together with turbidity, colour, chemical indicators, reservoir level and recent rainfall.

Deformation instruments include survey monuments, precise levelling points, extensometers, inclinometers, plumb lines, pendulums, GNSS receivers, tiltmeters and fibre-optic or robotic total-station systems. They measure horizontal and vertical movement, internal strain, rotation and differential displacement. Survey data require stable reference points; otherwise movement of the reference network can be misinterpreted as dam movement.

Stress and strain instruments include strain gauges, load cells, pressure cells, joint meters and crack meters. Their interpretation is sensitive to installation quality, temperature, creep, hysteresis and local rather than average behaviour.

Vibration and seismic instruments include accelerometers, strong-motion recorders, geophones and dynamic piezometers. They support assessment of earthquake response, resonance, machinery-induced vibration and structural impact. Vibration data require suitable sampling frequency, sensor orientation, calibration and signal processing.

Environmental and hydraulic instruments measure reservoir level, rainfall, temperature, inflow, outflow, gate position, spillway discharge and water quality. These are not merely auxiliary data: they are essential explanatory variables for interpreting structural response.

2.3 Instrument selection and network design

A professional design begins with a conceptual model of the dam and foundation. The engineer identifies zones of contrasting materials, drainage paths, potential slip surfaces, abutment contacts, joints, galleries and locations where failure-mode indicators would first appear. Instruments are then placed to test hypotheses about behaviour.

Important design criteria include:

• measurement range and resolution;
• accuracy, repeatability and drift;
• response time;
• survivability during construction and extreme events;
• redundancy for critical measurements;
• accessibility and maintainability;
• compatibility with telemetry and power systems; and
• cybersecurity and data ownership.

Redundancy does not mean installing identical instruments everywhere. It may involve independent measurement principles, such as comparing piezometer data with drainage discharge and visual observations.

3. Data Management and Interpretation

3.1 From reading to engineering information

A monitoring system should include a documented workflow:

1. acquire the reading;

2. record time, instrument status and loading conditions;

3. perform automatic and manual quality checks;

4. correct only for documented calibration or environmental effects;

5. compare with baseline behaviour and related instruments;

6. review trends and rates of change;

7. classify the condition; and

8. communicate and act according to the surveillance plan.

Quality assurance should identify missing values, impossible values, sudden offsets, sensor saturation, duplicated timestamps and changes caused by recalibration. Raw data must never be overwritten. Corrected data should retain the original value, correction method, person responsible and date of modification.

3.2 Baselines, thresholds and triggers

A baseline is a documented description of expected response under specified loading and operating conditions. It may be developed from design predictions, first-filling observations, historical records, numerical models and expert judgement.

Thresholds should be hierarchical:

• Normal or green: response is consistent with established behaviour.
• Alert or amber: response is unusual or uncertainty is increasing; verification and increased monitoring are required.
• Action or red: response may indicate an unsafe condition; responsible engineers and operators must implement predefined actions.

A threshold is not a universal constant. It depends on the instrument, season, reservoir level, rate of loading, uncertainty and failure mode. Rate-of-change thresholds are often more informative than absolute limits, particularly when progressive internal erosion or accelerating deformation is possible.

Statistical control limits can assist screening, but they must not replace engineering judgement. A simple trend model may be written as

\[

y(t)=a+bt

\]

where \(y(t)\) is a measured response, \(a\) is an intercept and \(b\) is the estimated rate of change. This model is useful only over a period in which loading and behaviour are approximately stable. Seasonal effects, reservoir-level changes, temperature and autocorrelation can make a simple linear trend misleading.

3.3 Piezometric interpretation

Piezometric data are interpreted by examining:

• hydraulic head relative to reservoir level;
• phreatic-surface position in an embankment;
• gradients toward drains and downstream slopes;
• spatial consistency between adjacent instruments;
• response time after reservoir changes or rainfall; and
• long-term trends.

An increase in pore pressure may be normal during reservoir filling, but a delayed or persistent increase can indicate clogged drains, changed seepage paths, cracking, internal erosion or altered boundary conditions. A sudden fall may indicate a broken instrument, a new drainage path or a genuine loss of confinement. The engineer must first confirm the instrument and compare neighbouring data.

3.4 Seepage interpretation

Seepage should be plotted against reservoir level and rainfall, not only against time. A rising seepage rate that is proportional to reservoir level may be expected, whereas an increase in turbidity, development of sinkholes, muddy discharge or a change in seepage location is more concerning. Seepage discharge should be separated, where possible, into drainage flows, leakage through joints and surface runoff to avoid mixing unrelated sources.

3.5 Deformation interpretation

Dam deformation is often reversible, seasonal or load-dependent. Concrete dams may move with temperature and reservoir loading; embankments may undergo settlement and creep. Interpretation should consider magnitude, direction, spatial distribution, reversibility and acceleration.

Differential movement is generally more informative than movement of a single point. A persistent change in displacement direction, opening of a joint, rapid settlement or correlation with increased pore pressure deserves investigation. Measurements from different systems should be compared, but apparent disagreement may result from different datums, frequencies, locations or measurement principles.

3.6 Vibration interpretation

Vibration monitoring can identify abnormal machinery operation, gate impacts, hydraulic instabilities or earthquake effects. Important characteristics include peak acceleration, dominant frequency, duration, damping and change from the established signature. A change in frequency may indicate stiffness loss, boundary-condition changes or altered water interaction, but frequency shifts can also result from sensor placement or operational changes. Signal interpretation therefore requires calibration records and knowledge of the excitation source.

4. Inspection and Surveillance

Instrumentation complements, rather than replaces, inspection. Routine inspections should examine upstream and downstream faces, crest, abutments, galleries, drains, spillway structures, gates, outlet works, access roads and reservoir margins. Inspectors should record location, dimensions, photographs, environmental conditions, water level and changes since the previous inspection.

Inspection quality improves when observations are categorised as:

• condition: what is physically observed;
• interpretation: what it may mean; and
• action: what must be done, by whom and by when.

Typical warning signs include wet areas, concentrated leakage, muddy water, sinkholes, longitudinal or transverse cracking, unusual settlement, bulging, erosion, displaced blocks, concrete deterioration, exposed reinforcement, gate malfunction and blocked drains. A warning sign does not establish a failure mechanism, but it should initiate verification and engineering assessment.

Surveillance frequencies should increase during first filling, extreme floods, earthquakes, rapid drawdown, unusual seepage, major repairs and abnormal instrument readings. Operators must have authority to report concerns without fear of blame or delay.

5. Failure-Mode Analysis and Risk-Informed Management

5.1 Failure modes

A failure mode is a physically plausible sequence leading from an initiating event to loss of a safety function and potentially to uncontrolled release. Examples include:

• internal erosion through an embankment or foundation;
• slope instability during high reservoir level or rapid drawdown;
• overtopping due to inadequate flood capacity or gate failure;
• sliding, overturning or excessive stress in a concrete dam;
• foundation dissolution or abutment instability;
• seismic cracking, liquefaction or deformation;
• spillway cavitation or erosion; and
• malfunction of outlet gates or power systems.

Failure-mode analysis should identify initiating loads, vulnerable features, intermediate indicators, barriers and consequences. A bow-tie representation is useful: threats lead to the failure event, preventive barriers act before the event, and mitigation barriers reduce consequences after the event.

5.2 Risk concepts

Risk is commonly conceptualised as a combination of likelihood and consequence. A qualitative formulation is

\[

R \sim P(F)\times C(F)

\]

where \(P(F)\) is the probability or likelihood of a failure scenario and \(C(F)\) is its consequence. The symbol \(\sim\) indicates a conceptual relationship, not an exact universal equation. Quantitative risk estimates require defined event frequencies, conditional probabilities, consequence modelling and uncertainty treatment.

Risk-informed management does not mean accepting any risk because it is numerically small. It means prioritising surveillance, investigation, maintenance and investment according to credible failure modes, uncertainty, consequence and practicable risk reduction.

5.3 Risk matrices and their limitations

Risk matrices can communicate priorities where data are limited. However, ordinal categories can conceal major differences within a category, exaggerate precision and encourage attention to colour rather than mechanism. A matrix should therefore be supported by failure-mode descriptions, evidence quality, uncertainty and explicit decision rules.

6. Operation, Emergency Planning and Response

Safe operation requires clear operating rules for reservoir levels, flood routing, gate movements, drawdown rates, inspection access, communication and abnormal conditions. Operators should understand not only what action is required but why it is required and what constraints apply.

An Emergency Action Plan (EAP) normally includes:

• conditions that activate the plan;
• responsibilities and authority;
• notification lists and redundant communication channels;
• reservoir and gate operating instructions;
• inundation maps and downstream warning arrangements;
• evacuation or protective-action information;
• equipment, access and power arrangements;
• contact with emergency services and regulators; and
• procedures for termination, documentation and post-event review.

Emergency levels should be based on observable conditions and credible consequences. Examples include confirmed unusual seepage, rapidly increasing deformation, spillway damage, forecast extreme inflow, earthquake damage or loss of gate control. Drills are necessary because plans that are not practised often fail at the interfaces between owner, operator, regulator, emergency services and communities.

During an incident, personnel should protect life first, establish reliable facts, avoid actions that worsen the condition, preserve records and communicate uncertainty honestly. Unauthorised public release of unverified information can create harm, but withholding credible warnings is ethically unacceptable.

7. Rehabilitation and Corrective Action

Corrective action should follow diagnosis rather than symptom suppression. A typical process is:

1. verify the observation and rule out instrument error;

2. inspect the affected area and review history;

3. identify plausible failure modes;

4. perform targeted investigation and analysis;

5. evaluate temporary risk-reduction measures;

6. design, review and implement rehabilitation;

7. monitor response and confirm performance; and

8. update drawings, operating rules and the safety case.

Possible measures include drain cleaning or replacement, filters and toe drains, grouting, cutoff walls, berms, slope flattening, anchors, concrete repair, spillway strengthening, gate replacement, erosion protection and improved power or communication systems. Repair may change seepage paths and therefore requires post-rehabilitation monitoring.

8. Digital Systems, Automation and Artificial Intelligence

Telemetry, cloud databases, remote sensing, distributed fibre-optic sensing, satellite interferometry and digital twins can improve coverage and response time. Machine-learning models may assist anomaly detection by identifying patterns in multivariable data that are difficult to observe manually.

These technologies have limitations. A model trained on normal behaviour may fail during unprecedented events; sensor drift may be classified as structural change; missing data can bias predictions; and opaque algorithms may not provide a defensible safety decision. AI should therefore support, not replace, engineering review. Systems should retain human override, uncertainty estimates, cybersecurity controls, audit trails and independent validation.

Digital transformation also creates governance questions: who owns the data, who can modify thresholds, how long records are retained, and who is accountable when an automated alert is ignored? These questions are part of dam safety, not merely information technology.

9. Governance, Professional Ethics and Regulation

The dam owner retains responsibility for providing resources, competent personnel, maintenance and emergency capability, even when activities are outsourced. Engineers have duties to public safety, technical integrity, confidentiality where legitimate, and truthful communication. Conflicts may arise when generation, irrigation or water-supply objectives encourage operation near safety limits.

A robust governance system includes an independent technical review function, documented competency requirements, controlled procedures, periodic audits, regulatory reporting and a learning culture. Near misses and anomalies should be investigated for system causes rather than concealed or attributed automatically to individual error.

Regulatory requirements differ among jurisdictions, but commonly address inspection frequency, design review, instrumentation, operation, emergency planning, incident reporting and consequences classification. Engineers must follow applicable national standards and regulatory directions; this chapter provides principles rather than a substitute for local law.

10. Illustrative Case: Rising Seepage at an Embankment Dam

Background

An illustrative 42 m-high zoned earthfill dam has operated for 18 years. It contains foundation piezometers, downstream toe drains, V-notch weirs and survey monuments. The reservoir is seasonally operated.

Problem

During a high reservoir period, one toe-drain weir records discharge 35% above its established relationship with reservoir level. Two nearby piezometers show increasing head, while a third instrument shows no change. The discharge appears slightly turbid after heavy rainfall.

Analysis

The initial response is not to declare failure. The owner verifies the weir calibration, checks for blockage and confirms reservoir and rainfall records. A site inspection finds no sinkhole or visible downstream wet area, but a drain outlet is partially obstructed. After clearing the obstruction, the measured discharge changes but remains above the historical envelope. Independent flow measurement confirms the abnormality.

The engineering team develops possible explanations: increased rainfall infiltration, drain-system alteration, instrument error, or a developing concentrated seepage path. The spatial pattern of two rising piezometers and persistent turbidity increases concern about a local seepage change. Failure-mode analysis identifies internal erosion as a credible mechanism, although evidence is not yet conclusive.

Approach and solution

The dam is placed under an alert condition. Reading frequency is increased, a downstream inspection is conducted daily, turbidity is measured, and temporary piezometers are installed. Reservoir operations are adjusted to avoid further rise while the investigation proceeds. Borehole exploration and geophysical surveys identify a coarse zone near the foundation contact. Engineers install a filtered relief-drain system and reconstruct part of the drainage collection system under controlled conditions. The emergency plan is reviewed with local authorities but is not fully activated because no rapid deterioration is observed.

Outcome

Following rehabilitation, piezometric heads decline toward the previous operating envelope and seepage becomes clear and stable. Continued monitoring demonstrates that the response is consistent across independent instruments. The owner updates the failure-mode register, adds a turbidity trigger, improves drain access and revises maintenance procedures.

Learning points

The case demonstrates the value of combining instruments rather than relying on one reading; verifying data before escalating; treating turbidity and changing seepage location as important indicators; using staged operational controls; and linking corrective work to a specific failure mode. It also shows that alert-level management can be prudent even when failure has not been proven.

Challenges and Limitations

Dam safety decisions are affected by ageing instruments, incomplete construction records, changing climate, nonstationary flood and rainfall patterns, uncertain seismic hazards, limited budgets and loss of experienced personnel. Automated systems may create false alarms, while poorly designed thresholds may create complacency. Numerical models can provide useful insight but remain dependent on assumptions about material properties, boundary conditions and failure mechanisms.

Climate change may alter design floods, drought-related drawdown cycles, sedimentation, temperature effects and downstream exposure. Asset managers should periodically reassess these factors rather than treating historical records as stationary indefinitely. The most important limitation is often organisational: data may exist but not reach a person with the authority and competence to act.

Best-Practice Framework

A mature dam safety programme should:

1. maintain a current conceptual and numerical understanding of the dam;

2. link every critical instrument to a failure mode and decision;

3. preserve raw data and document validation and corrections;

4. use multiple independent evidence sources;

5. define alert and action thresholds with response procedures;

6. inspect systematically and record changes spatially;

7. review hydrology, seismicity, climate and downstream consequences periodically;

8. test emergency plans through exercises;

9. provide independent technical review and adequate resources; and

10. learn from incidents, near misses and rehabilitation outcomes.

Chapter Summary

Dam safety management is a lifecycle system that integrates loading, measured response, inspection, engineering interpretation and operational decision-making. Instrumentation includes piezometers, seepage meters, deformation and strain devices, vibration sensors, and environmental and hydraulic monitors. Their value depends on appropriate selection, calibration, redundancy, reliable data governance and interpretation against loading conditions and established baselines.

Piezometric, seepage, deformation and vibration data should be assessed through trends, spatial relationships, rates of change and independent corroboration. Thresholds are decision tools, not universal proofs of safety. Failure-mode analysis identifies credible physical sequences, barriers and indicators, while risk-informed management prioritises action according to likelihood, consequence and uncertainty.

Effective programmes also require competent operation, inspection, rehabilitation, emergency action planning, ethical governance and regulatory compliance. Digital and AI technologies can improve detection but must remain auditable, validated and under accountable professional oversight. Ultimately, dam safety is achieved not by instruments alone but by a prepared organisation capable of recognising evidence and acting on it.

Key Terms

Alert threshold: A predefined condition requiring verification or increased surveillance.

Baseline behaviour: Documented normal response under specified loading and operating conditions.

Dam safety: Management of conditions and processes that protect the dam and downstream population from uncontrolled release.

Emergency Action Plan: A documented plan for warning, coordination and response during a dam-related emergency.

Failure mode: A physically plausible sequence leading to loss of a safety function.

Foundation: Natural or constructed ground supporting and interacting with the dam.

Hydraulic head: Elevation plus pressure head at a point in a hydraulic system.

Instrumentation: Devices and systems used to measure dam, foundation and loading responses.

Internal erosion: Detachment and transport of soil particles by seepage.

Piezometer: An instrument used to measure pore-water pressure or hydraulic head.

Phreatic surface: The surface within an embankment below which soil pores are substantially water-filled.

Risk-informed management: Allocation of attention and resources using failure likelihood, consequences and uncertainty.

Seepage: Movement of water through or beneath a dam and its foundation.

Surveillance: Systematic inspection, monitoring, interpretation and reporting of dam condition.

Telemetry: Remote transmission of measured data to a receiving system.

Threshold: A predefined measurement or condition associated with a management response.

Turbidity: Cloudiness of water caused by suspended particles.

V-notch weir: A calibrated triangular opening used to measure flow discharge.

Review Questions

Short Answer Questions

1. What are the three principal elements linking dam monitoring to safety decisions?

2. Distinguish pressure head from total hydraulic head.

3. Why should seepage discharge be interpreted together with turbidity and reservoir level?

4. State two limitations of automated or AI-based dam monitoring.

5. What is the purpose of an Emergency Action Plan?

Long Answer Questions

1. Explain how a monitoring network should be designed from identified failure modes.

2. Discuss the interpretation of piezometric, deformation and seepage data during abnormal conditions.

3. Describe the lifecycle components of a dam safety management system.

4. Explain how failure-mode analysis supports risk-informed rehabilitation.

5. Evaluate the roles of inspection, instrumentation, governance and independent technical review in dam safety.

Analytical / Application Questions

1. A toe-drain discharge rises sharply while reservoir level remains constant, but only one of three nearby piezometers changes. Develop a verification and response sequence, including possible explanations.

2. A concrete dam shows a seasonal crest displacement that is reversible, while one joint meter shows progressive opening. Explain how you would distinguish normal thermal behaviour from a developing safety concern.

3. An owner proposes reducing inspection and instrumentation budgets because no threshold has been exceeded for five years. Prepare a risk-informed argument for or against the proposal, considering uncertainty, ageing, consequences and organisational resilience.

Multiple Choice Questions with Answers

Question 1

Which measurement most directly indicates pore-water pressure within an embankment or foundation?

A. Piezometer reading

B. Survey monument position

C. Rainfall depth

D. Gate opening percentage

Correct Answer: A. Piezometer reading

Question 2

The total hydraulic head at a point is expressed as:

A. z − u/γw

B. z + u/γw

C. uγw + z

D. u/z

Correct Answer: B. z + u/γw

Question 3

A sudden increase in seepage accompanied by muddy discharge is most concerning because it may indicate:

A. Normal thermal expansion

B. Internal erosion

C. Reduced reservoir evaporation

D. Survey-network movement

Correct Answer: B. Internal erosion

Question 4

What is the first appropriate response to an unexpected instrument reading?

A. Immediately declare catastrophic failure

B. Delete the reading as an outlier

C. Verify the instrument and compare related evidence

D. Increase the reservoir level to test it

Correct Answer: C. Verify the instrument and compare related evidence

Question 5

Which item is an example of a preventive barrier in a bow-tie failure analysis?

A. Downstream evacuation after a breach

B. A functioning filter designed to prevent internal erosion

C. Post-incident media reporting

D. A historical photograph

Correct Answer: B. A functioning filter designed to prevent internal erosion

Question 6

Why are stable reference points necessary for deformation surveys?

A. To increase reservoir storage

B. To distinguish dam movement from reference-network movement

C. To eliminate all temperature effects

D. To measure turbidity

Correct Answer: B. To distinguish dam movement from reference-network movement

Question 7

A risk-informed dam safety programme primarily prioritises actions according to:

A. Instrument brand and purchase price

B. Failure likelihood, consequences and uncertainty

C. The age of the operator

D. The number of available photographs

Correct Answer: B. Failure likelihood, consequences and uncertainty

Question 8

Which condition commonly warrants increased surveillance?

A. Routine stable operation with no changes

B. An earthquake affecting the dam region

C. A documented unchanged baseline

D. A completed data backup

Correct Answer: B. An earthquake affecting the dam region

Question 9

What is a major limitation of applying machine learning to dam monitoring?

A. It cannot store numerical data

B. It may misclassify unprecedented events or sensor drift

C. It eliminates the need for sensors

D. It always produces physically causal explanations

Correct Answer: B. It may misclassify unprecedented events or sensor drift

Question 10

The principal purpose of an Emergency Action Plan is to:

A. Replace all routine inspections

B. Define warning, coordination and response procedures during emergencies

C. Set construction material prices

D. Calculate daily evaporation only

Correct Answer: B. Define warning, coordination and response procedures during emergencies