Tree Risk Assessment,
Explained

A tree inventory is an asset register — species, size, condition, location, for every tree on a property. A condition rating (good, fair, poor, dead) is one field inside that register, a general health snapshot.

A tree risk assessment is a different, more specific thing: a dated evaluation of whether a particular tree, or part of it, is likely to fail and what happens if it does. ISA's Tree Risk Assessment Qualification (TRAQ) is the credential most US practitioners hold to perform this kind of assessment, and its methodology — likelihood of failure, likelihood of impacting a target, consequence of failure, combined into an overall risk rating — is the framing most reports and most insurers and municipalities now expect to see, whether or not the assessor is formally TRAQ-qualified.

A tree can have a "Fair" condition rating and still carry a "High" or "Extreme" risk rating, if the defect that's dragging its condition down sits directly over a target. Condition and risk answer different questions.

Not every tree gets the same depth of inspection, and matching the level to the situation is itself part of the professional judgement. The TRAQ framework distinguishes three:

Risk without a target isn't risk in the sense that matters for a report — a tree could shed its entire canopy tomorrow in the middle of an unused field and it's a non-event. The target is whatever's in the fall zone: a house, a driveway, a footpath, a parked car, a play area, a property line, people.

Record the target specifically. "Rear patio, used daily in summer" is a useful record. "Nearby structure" is not — six months later, neither you nor anyone reading the report will remember which structure, or how often anyone stands under it.

A judgement of whether the tree, stem, branch, or root system in question is likely to fail within the assessment's time frame, based on the defects observed: decay, cracks, cavities, included bark, lean, root damage, deadwood, pest or disease activity. The standard TRAQ matrix scores this as improbable, possible, probable, or imminent.

This is a judgement call informed by observation, not a formula. Two arborists looking at the same crack can reasonably land on "possible" versus "probable" — which is exactly why documenting what was actually observed (species, location, extent, photos) matters more than the label alone.

If the tree (or the part in question) does fail, what happens to the target? Scored as negligible, minor, significant, or severe — driven by the size of what could fail, the target's value and occupancy, and how much of the target is actually exposed. A large stem over a busy footpath scores very differently from the same stem over an unused corner of a field.

Occupancy matters as much as proximity. A target that's occupied constantly (a public footpath, a busy car park) carries higher consequence than one occupied rarely (a fenced storage yard visited once a week), even if the physical distance from the tree is identical. Record occupancy, not just presence, when you note the target.

Likelihood and consequence combine into an overall risk rating: low, moderate, high, or extreme. The rating drives what happens next — low risk usually means monitor and reinspect on a normal cycle; extreme risk usually means immediate action, sometimes before you've even left the site.

The rating is a summary, not the deliverable. A client, insurer, or court doesn't just need "High" written on a page — they need the observations and reasoning behind it, in case the rating is ever challenged.

As a rough shape: a tree with a probable or imminent likelihood over a target with significant or severe consequence lands high or extreme, regardless of how minor either factor looks in isolation. A tree with a possible likelihood but negligible consequence — because there's genuinely nothing valuable in the fall zone — stays low even with a real defect present. The rating isn't a straight average of the two inputs; a severe consequence pulls the rating up even against a modest likelihood, because the cost of being wrong is asymmetric.

Every risk rating needs a next step. Mitigation is what you recommend to bring the risk down — prune out the specific defect, cable or brace a co-dominant union, remove the tree, restrict access to the target area, or simply monitor. Write the mitigation as an action, not a vague sentiment: "reduce deadwood over 50mm diameter in the crown" is actionable; "improve tree health" is not.

Set a reinspection date. Risk assessments have a shelf life — a tree rated "Low" in dry summer conditions can look different after a wet winter and root plate movement. A reinspection date on file is also part of your professional defensibility: it shows the assessment wasn't a one-time look that then got forgotten.

How you set the interval depends on the rating and the target. A low-risk tree over a rarely used corner of a property might go on a normal multi-year inspection cycle. A moderate or high-risk tree over a frequently occupied target usually needs a much shorter interval — months, not years — until either the defect is addressed or the tree is removed. Extreme-risk findings typically call for immediate mitigation rather than a scheduled reinspection at all.

The real value of a risk assessment usually only surfaces after something goes wrong — a limb comes down, a tree fails, and someone asks whether it was inspected, when, by whom, and what was recommended. At that point the rating you assigned matters far less than whether you can produce a dated record: what you observed, what target you identified, what you recommended, and whether the recommendation was acted on.

A risk assessment that lives in your memory or a paper notebook that's since been thrown out is not a defensible record. A risk assessment tied to a tree number, a date, and photos is.

LogLog records the same fields this guide walks through, against the same tree pin you already dropped on the site plan: failure likelihood (improbable, possible, probable, imminent), failure consequence (negligible, minor, significant, severe), an overall risk rating (low, moderate, high, extreme), the target, a mitigation note, and a reinspection date. All of it exports with the rest of the tree record — it's included in the ZIP regardless of which fields you've chosen to show in the editor, so a risk observation never quietly gets left out of a report.

LogLog doesn't assign the rating for you. The likelihood, consequence, and mitigation call is still the professional judgement of whoever is standing at the tree — LogLog just makes sure that judgement gets written down, dated, and photographed before you've moved on to the next tree.