Monday morning, about 5:30 on Dec. 21, I heard a sound I had not heard for a year: water gushing from the outlet of the sump pump under our house.

The pump is triggered by runoff when the subsoil is saturated. It is a reminder we are in a time of increased likelihood of soil/tree failures, e.g., soil and tree roots pulling apart.

Notice I say “likelihood,” not “risk” or “hazard.” There are important differences in word meaning when we are talking about tree risk assessment. More about that in a minute.

First, what keeps a tree standing up? Sound wood, a well-developed root system, and sound soil.

Sound wood in the trunk and buttress roots has great compressive and considerably less tensile strength. Sound roots have great tensile strength, like ropes, cords, strings and threads intertwined with the soil.

Soil has great compressive strength — it can bear heavy loads pushing downward on it — but far less tensile strength. It pulls apart easily, especially when wet.

Many of the tree failures I have seen have been of this variety. Saturation lubricates the interfaces of soil particles and the interfaces of roots with soil particles. The anchorage system unravels when it is pulled apart by weight and wind.

In some cases, shallow, flat root systems tip upward, like the base of a fallen wine glass. That happens when wind forces cause roots to lose their grip on wet soil, especially where deep “sinker” and “heart” roots are lacking in shallow or compacted soils.

There was a great example lying along the side of Highway 12 near Glen Ellen. I don’t know if it is still there; a gigantic Eucalyptus root plate tipped up to vertical, revealing no intact deep roots.

In other cases, the entire root mass rotates in wet soil. One side lifts up while the opposite side sinks down. I have seen this with old valley oaks in deep clay, particularly in over-mature trees subjected to saturated soil and strong, gusty winds.

Now, about “hazard,” “likelihood” and “risk,” we need to pay attention to the trees and potential targets, but also the language we use. This is covered nicely in the current system, described in Tree Risk Assessment (International Society of Arboriculture, BMP — Best Management Practices, 2011). Arborists who complete the course are Tree Risk Assessment Qualified (TRAQ).

Under the current system, risk is the “combination of the likelihood of an event and the severity of the potential consequences.” A hazard is “the tree part(s) identified as a likely source of harm.”

The heart of the system goes like this. The arborist determines:

— The likelihood of a tree failure (that’s a professional word for falling over or falling apart). The likelihood can be “Imminent,” “Probable,” “Possible” or “Improbable” within a time frame such as one year or whatever time frame the owner chooses.

Get home and garden tips sent to your email inbox

— The likelihood of impacting a target, which can be whatever the tree owner is concerned about. Likelihood can be “Very Low,” “Low,” “Medium” or “High.”

— The consequences in the event of failure and impact.

Following these steps through the risk-assessment matrix given in the BMP, the arborist arrives at a risk rating of “Extreme,” “High,” “Moderate” or “Low.” The only way to achieve zero risk is to have zero tree, or for it to be out of range of the potential target of concern.

It is the responsibility of the tree owner or manager — not the arborist — to make a decision and choose a course of action.

A level of risk unacceptable to the owner should be mitigated to an acceptable level, typically to achieve a “Low” level of “residual risk” for a chosen time frame.

In a nutshell, that is what TRAQ Arborists are using nowadays, and what the profession and the American National Standards Institute prescribe.

Bill Pramuk is a registered consulting arborist. Visit his website, Email questions to or call him at (707) 226-2884.