How to track a crack that changes over time

Give each crack an identifier and record its location, date, overall and close-up photos, repeatable measuring points, initial reading, and any moisture, spalling or offset. Compare a series, not two photographs taken in different ways.

This guide goes deeper into monitoring. If you do not yet know whether the crack is in a finish, block, column or beam, start with our guide to deciding when an assessment is needed. It also offers a free crack log that you can keep using. Here we explain how to make entries comparable and where their limits lie. A residential study by the Universidad Tecnológica de Panamá (UTP) began with a preliminary inspection and a damage record to establish a baseline before monitoring. Source: UTP, 2023, p. 48

Stop if there is danger. A damaged column or beam, a leaning wall, falling material or an apparent collapse risk calls for action, not weekly photographs. Move away and report the condition. SINAPROC advises staying out of structures with visible damage or collapse risk and reporting cracks, leaning and deterioration. If a home was evacuated after an earthquake, Panama's Fire Department advises against returning until it has been assessed. Sources: SINAPROC, 2026; Fire Department, 2025

Monitoring makes sense only when you can observe the area safely and the immediate safety question has been resolved.

• Emergency or restricted area: do not go in to install a tell-tale or photograph a crack. Describe what you can see from a safe place and follow responders' instructions. Sources: SINAPROC; Fire Department
• New damage in a possible load-bearing member: if the crack is in a column, beam, slab or wall whose role you do not know, request an assessment before patching it or waiting weeks. When a formal structural evaluation is commissioned, Panama's REP-2021 examines the actual existing structure with field verification. Source: REP-2021, §§15.1 and 15.3.1, pp. 102 and 105
• Change with other signs: spalling, an offset between sides, new seepage or visible distortion should be recorded and communicated, not turned into a diagnosis by the homeowner. USACE includes these associated conditions in a professional crack survey. Source: USACE, p. 2-11

If the damage appeared after an earthquake, use the reentry guide to address access first. The date of discovery does not prove the cause: in August 2026 inspections, SINAPROC distinguished possibly preexisting conditions from possible earthquake effects. Source: SINAPROC, August 2026

The log can answer, “What changed, where, and under what conditions?” It cannot answer on its own, “What caused it?” or “Is the house safe?”

Keeping these questions separate prevents a premature conclusion. A series can show whether an opening at one point increases, decreases or appears to cycle; a sketch can show whether the line lengthens or new cracks appear nearby. But a reading depends on location, method and instrument. The U.S. National Park Service describes daily and seasonal movement related to temperature and moisture in historic masonry, and warns that a short series can be misleading. That document does not establish a pattern for a Panamanian home. It shows why two readings alone cannot establish a cause. Source: NPS, pp. 43-44

In a Panamanian case, UTP researchers combined crack gauges, comparators, geometry checks, radar and vibration analysis. Their conclusion about the cause came from that combined investigation, not from one photograph or number. Source: UTP, 2023, pp. 49 and 53-54

If a door also began to rub or stopped closing, record that date separately and use the door and cracks guide to decide when tracking the crack alone is insufficient.

The first visit establishes the reference. If you cannot identify the same point next time, the comparison loses value.

1. Give each crack an identifier: for example, L1 for the living-room wall and L2 for another near the window. This is your own convention, not an official code. Do not combine readings from different cracks.
2. Draw a sketch: note the room, wall, nearby door or window, floor and face observed. If the crack crosses a joint, show the joint. FEMA 306 calls for the wall's location, crack orientation and visible damage on a sketch in a professional inspection. Source: FEMA 306, p. 31
3. Photograph context and detail: take an overall view that locates the crack, then a close view with a ruler alongside if the place is safe. Keep roughly the same camera position and note changes in lighting. A crack highlighted with a marker cannot be sized from the image. Source: FEMA 306, pp. 31-32
4. Select repeatable points: mark points A, B and C on the sketch, away from chipped edges. Width can vary along a crack, so a single measurement does not describe the whole line. Source: FEMA 306, pp. 31-32
5. Record the initial condition: date, time, first reading, tool and any moisture, loose material or offset. Do not remove finishes or open a wall to discover the depth. Source: USACE, p. 2-11

If you have an earlier report or photographs taken before an earthquake or previous repair, keep their original dates. SINAPROC has noted that older damage must sometimes be distinguished from new effects. An earlier photograph can help frame that question, although it cannot settle it alone. Source: SINAPROC, 2026

A ruler helps locate a crack and measure its length. A crack comparator estimates the opening at a point. An installed gauge follows relative movement. These are different readings.

If you use a comparator, place it at the same point and in a similar orientation each time. Do not read across an edge that broke off between visits: the reading may change even if the body of the wall did not open by the same amount. Record the tool and keep the photograph. FEMA 306 calls for representative points away from holes or chipped edges and notes that lighting can cause a plastic card to overestimate width. These are limits of the method, not permission to assess structural capacity yourself. Source: FEMA 306, pp. 31-32

If a professional installs a gauge across the crack, ask for the actual initial reading. USACE warns that it may not read zero when installed. Later readings are interpreted against that baseline, with the movement direction and location documented. Pencil marks at the ends may help show whether the line grows longer, but they do not measure displacement between its two sides. Source: USACE, p. 2-11

Avoid reporting decimal places your tool or photograph cannot resolve. If two photographs or readings cannot be compared, write “not comparable” and explain why. Visible uncertainty is more useful than a false trend.

Use the least intrusive observation that answers the question. Installing a device or changing the surface may affect evidence and access.

UTP used several instruments in an occupied residential building, along with checks of distortion and concealed components. That does not mean every home needs nine sensors, radar or six months of waiting. The scope and plan should fit the damage and the decision to be made. Source: UTP, 2023, pp. 48-49

These sources do not prescribe one interval for every crack. Review frequency and triggers depend on the problem and, when a professional is involved, the agreed plan.

If the area is safe and you are making preliminary observations, choose a day and time you can repeat and label each entry consistently. Add an extra observation after an event that may matter, such as heavy rain, seepage, renovation or felt shaking, but never enter a restricted area to do so. Distinguish the event date from the date you first saw the crack. SINAPROC has warned that damage discovered after an earthquake may predate it. Source: SINAPROC, 2026

Record conditions you can actually observe: recent rain, a wet wall, a repair, loose pieces or a door that has begun to rub. Do not write “rain caused the crack” merely because both happened that week. The NPS explains that some cracks in historic masonry open and close with environmental cycles, which calls for a series and context before assigning a trend. That foreign example supports caution; it does not establish a Panama monitoring calendar. Source: NPS, p. 43

If an engineer is already responsible, ask in writing which points to monitor, which instrument to use, how often to read it and which change requires a call before the next scheduled review. Do not borrow alert thresholds from another building or a commercial page.

UTP followed persistent cracks in the walls of an occupied residential building. It is evidence of an investigation method, not a rule for houses.

The team established a baseline, used nonintrusive methods over six months and installed nine devices on four walls. Only four instruments recorded crack movements of 0.2 mm or less; for that building, the researchers described no relevant movement. They also checked geometry, slabs, concealed wall components and vibration response. Source: UTP, Méndez-Serrano et al., 2023, pp. 48-51

The conclusion was specific to that building: the cracking in the masonry elements was not explained by excessive behavior of the assessed slab system; the team related it to insufficient confinement details in those elements. Neither the cause nor the 0.2 mm reading can be copied to another property. The useful lesson for a homeowner is that a trend is interpreted alongside the construction system and other checks. Source: UTP, 2023, pp. 53-54

If someone offers to “measure once and certify,” ask which baseline conditions, members and limitations they will examine. For a formal structural evaluation in Panama, REP-2021 calls for a model of the actual structure and a field study. A homeowner's log can inform that work, not replace it. Source: REP-2021, §15.3.1, p. 105

Before saying “it grew,” check that each entry uses the same point, tool, wall face and reading method.

Avoid plotting uncertain points and announcing a “growth rate.” The NPS warns that short series may hide cycles; USACE reminds readers that a gauge's initial reading is not always zero. A well-located series with notes on changing conditions is often more useful than a precise-looking number without context. Sources: NPS, p. 43; USACE, p. 2-11

These examples are fictional and show how to compare records. Their numbers are not safety thresholds or suggested waiting periods.

1. Example 1: comparable measurements. Mark L1-A in the plaster of an accessible wall was recorded as 0.8 mm on October 12 and 1.0 mm on October 19, using the same comparator and point. The arithmetic difference is 0.2 mm. The file contains two overall photographs, two details and the tool used. It makes no claim of “severe damage” or “safety.” The trend is reported if it persists or the crack is in a supporting member. FEMA 306 explains the importance of representative points and measurement conditions. Source: FEMA 306, pp. 31-32
2. Example 2: not comparable. The first picture was close with side lighting; the second was taken from a hallway without a scale. The crack appears wider, but the images offer no measurement basis. Keep both, explain the limitation and take a new baseline only if access is safe. A line added with a marker can locate a crack but cannot establish its width in a picture. Source: FEMA 306, p. 31
3. Example 3: monitoring stops. A family has been logging a finish crack. One day they see fallen material and a visibly leaning wall. They do not wait for the monthly reading or install a tell-tale. They leave and report the signs; the earlier record can be provided later to the assessment team. Source: SINAPROC, 2026

An organized file lets the professional see the sequence before visiting, but does not replace the visit.

Provide a sketch with each crack identifier, dated original photographs, measuring points, baseline reading, tool, complete series and notes about repairs, rain or earthquakes. Include older pictures if available. Do not edit the only original to “make the crack visible.” Explain which areas you could not see because of danger or lack of access. For a formal evaluation, REP-2021 treats existing plans as preliminary reference and calls for field checks of the actual building and damage. Source: REP-2021, §15.3.1, p. 105

Ask the qualified professional to state in writing how they will use the series, whether instruments will be installed, how long they will be reviewed, who receives alerts and what they will deliver. Our guide to commissioning an inspection explains scope and reports; the JTIA public search lets you check professional registration. Source: JTIA, professional search

Do not conceal earlier repairs. SINAPROC has documented cases where potentially preexisting damage had to be distinguished from new effects. A chronology matters precisely because it may show that a crack existed before an event, or that its first appearance is unknown. Source: SINAPROC, 2026

If the chronology may also support a claim, keep the original record and read the policy without treating the crack date as automatic proof of cause. The home insurance and earthquake guide separates the technical record from notice to the insurer.

Share the chronology with whoever coordinates the assessment, without assigning a cause or responsibility yourself.

Under Ley 284, columns, beams and slabs are essential common property in a PH condominium. If the crack is in one of these members or an area affecting other units, notify management with its location and photos taken safely. Ask which areas are restricted and who will inspect. A log from one unit does not necessarily describe the whole building. When the origin and economic amount of damage between property or units must be established, article 29 provides for a qualified expert. Source: Ley 284 de 2022, arts. 16 and 29

After an earthquake, prioritize official access instructions and the reentry assessment. The Fire Department advises against returning until the home has been evaluated. If access is restricted, do not go inside to finish a table. Record what you already know from a safe place and keep older documents: SINAPROC has noted that some cracks observed during post-earthquake inspections may predate the event. Sources: Fire Department, 2025; SINAPROC, 2026

• Comparing one close-up with an overall picture without knowing the exact point. Keep both scales and a sketch. FEMA 306, p. 31
• Reading a gauge as if it began at zero. The initial reading is essential. USACE, p. 2-11
• Measuring across a broken edge and calling the difference an opening of the whole wall. FEMA 306, p. 31
• Deciding from two readings that a crack is “dead” or that rain or shaking caused it. Movement may be cyclical; discovery date does not prove causation. NPS, p. 43; SINAPROC, 2026
• Covering or painting the crack, or installing a tell-tale, before a possible structural problem is documented. Keep the evidence and let the professional decide what to open or change. REP-2021, §15.3.1
• Continuing to log an area where pieces are falling, a wall is leaning or collapse appears possible. Safety comes before the log. SINAPROC

Accessed October 10, 2026. The full UTP study and NPS publication were read in their original PDFs; USACE, FEMA, REP and Ley 284 were checked against their original pages.

Notice. This log does not certify a home as safe, establish a cause or replace emergency instructions or an on-site assessment. The numerical examples on this page are fictional and show only how to compare records. Updated October 2026.