Expert: Grizzly Peak Wildfire Response Reveals a Potential Flaw

By Glen Martin

Last week’s wild fire on Grizzly Peak Boulevard ended up scorching about 20 acres of brush and grass near the Lawrence Berkeley National Laboratory, with no major damage to property and no loss of life. That was due in large part to a fuel reduction program pursued by Berkeley Lab since the 1990s, says Scott Stephens, a professor with UC Berkeley’s Department of Environmental Science, Policy, and Management and one of the country’s foremost wildfire experts.

“The lab has done a good job of treating blue gum [eucalyptus] stands around their facilities and generally removing or reducing fuels,” says Stephens, noting the ubiquitous eucalypts that grow across the East bay hills are notorious for their flammability. Due at least partially to Berkeley Lab’s efforts, says Stephens, the fire never “crowned”—that is, it didn’t leap up the “fuel ladders” of brush and branches to the interconnected canopies of the trees, then rage across the landscape.

Firefighters also were able to jump on the flames in short order, due in large part to a signage system put in place by Tom Klatt, Stephens’ predecessor at Berkeley.

“I try not to promote draconian scenarios, but I am concerned about them,” Stephens says. “A fire driven by a strong east wind on a hot day would’ve acted very differently.”

“Tom was concerned about just this kind of situation: a fire breaking out on Grizzly Peak Boulevard, with firefighters delayed because there are so many twists and turns to the road and it can be difficult to determine location,” Stephens said. “So he was able to get a series of sign posts installed up there. Last week’s fire started at sign post 14, and firefighters were able to respond very rapidly because they knew exactly where they were going.”

But the East Bay Hills dodged an incendiary bullet for another reason, Stephens says: On the day the fire ignited, the weather was mild and a west wind was blowing, more or less pushing flames away from Berkeley Lab and the UC Berkeley campus. If the fire had started on a hot day with an east wind—the conditions that prevailed during the disastrous Oakland Hills fire of 1991—things might have concluded tragically.

“I try not to promote draconian scenarios, but I am concerned about them,” Stephens says. “A fire driven by a strong east wind on a hot day would’ve acted very differently. It not only would’ve burned very quickly, but where particularly volatile fuels such as eucalyptus are concerned, it would have thrown embers miles ahead, starting hundreds of spot fires that would also burn explosively and merge. That’s what happened in 1991.”

Normally, wildfires burn more rapidly uphill than downhill, observes Stephens, but in extreme conditions such as those that characterized the Oakland Hills Fire, “the fire overwhelms the topography. If last week’s fire had occurred under Oakland Hills fire conditions, there would’ve been impacts to university property. I’m particularly concerned about the Clark Kerr campus dormitories. They seem at significant risk.”

Stephens has long supported major fuel reduction programs for the East Bay Hills, and was particularly distressed when FEMA pulled funding for a plan to remove eucalyptus and restore native vegetation on land owned by the university along Claremont Canyon. But fuel reduction, he avers, is not enough. Public facilities in the East Bay Hills also need effective response programs. Berkeley Lab evacuated its employees during the Grizzly Peak fire, says Stephens, “and I watched from the campanile for about 45 minutes as cars slowly snaked down the hill. People were particularly vulnerable at that time. If the fire had burned down that road rapidly, as you could expect with a strong east wind, they would’ve been in trouble. The natural instinct for people surrounded by wildfire is to get out of their cars and run, which can put them at even greater risk. Fatalities occurred during the Oakland Hills Fire because of that reaction, and it happens regularly during Australian wildfires.”

Many of Berkeley Lab’s buildings are heavy concrete structures that could be largely impervious to wildfire, particularly given the aggressive fuel reduction policies the lab has pursued for the past 20 years, Stephens observes.

“I think it might be time for them to consider a shelter-in-place policy for their wildfire response strategy,” he says. “It would warrant an evaluation, at least. And they might also consider additional entrances and exits, a major reason the cars were so slow in getting out is because access is so limited. I also think UC needs to revisit and update its evacuation policies for its buildings, including the Clark Kerr dorms.”

Share this article:
Google+ Reddit


Here is a very different perspective on last week’s fire in the East Bay Hills: The fire started on the west side of Grizzly Peak in an area in which 1,900 eucalypts were clear cut by UCB in 2004. The logs of the trees that were destroyed in 2004 lined the road and burned in the fire. The fire spread across the road into Tilden Park where the first STOPPED. The eucalyptus trees in Tilden Park did NOT burn. Professor Stephens’ opinions about fire hazards are rather contradictory. In this interview, he blames the fires this season on the grasses that were encouraged to grow by heavy rains: He also says that forests are less likely to burn than grass because they are moist: “UC Berkeley Fire Science Professor Scott Stephens says most of the fires so far have been in grassland areas that were revived from the rain, then dried out early during triple-digit heat waves… He says forests are better at retaining moisture and the Sierra will be more resilient this year because of the rains.” Those who use their eyes to evaluate the fire will see through the smokescreen of those who demand the destruction of our urban forest.
One has to wonder whether Professor Stephens actually visited the site of this fire before providing his expert assessment. What happened at this site was the the majority of this fire occurred on UC land that HAD BEEN TREATED by UC several years back. Did this treatment prevent a fire? Absolutely not. Did this fire burn voraciously? Yes. At the northern end of the site was an area that had not been treated by UC. As a result of the intact tree canopy there were fewer ground fuels, but the ground fuels that were there did burn. But the eucs and pines at the site DID NOT BURN. On the other side of the street is an EBRPD site where there is a euc grove with aggressively managed understory. While the fire spread to the area just to the north of this site, where the understory had not been managed, there was ZERO SPREAD of the fire in the euc grove with managed understory. This is what’s called a shaded fuel break. It worked. While Professor Stephens failed to mention these details, he did opine on how much worse this could have been had the winds been stronger. This is no doubt true, but the same basic dynamics apply. Where there are understory fuels there will be fire and where there are little or none it’s highly unlikely there will be fire. As to concern about crown fires one can’t speak of this risk in isolation. i.e.. while a crown fire in a euc grove would be difficult to manage, the far greater risk is a crown fire in vegetation with a high percentage of “fine fuels” (<3” in diameter) and with crown at or near ground level. A prime example of such a species is the bay tree. Other agencies in this area have recognized this threat and are limbing up bay trees to eliminate this hazard. While Professor Stephens might argue that what happened last week was anomalous, the reality is that it wasn’t. We’ve toured every wild land fire in this area since ’91 and the results have been nearly identical. All the ground fuels burn and few if any of the tall trees burn. Eucs seem particularly resistant to ignition due to their high moisture content and the fact that their crowns are very high above ground level. One might almost get the idea that they evolved to be resistant to fire. As to Tom Klatt’s role in putting up signs, yes this is a good thing. But the bigger issue is that ever since UC cleared out all the tall trees that were blocking views and expanded the pullouts along Grizzly Peak there have been a series of fires started by revelers who have been using these facilities. We think it high time that UC accept responsibility for the increased risk and create vegetation free zones around these pullouts so that what happened last week is less likely to happen in the future. In short, as Mary McAllister wrote, the reality on the ground simply doesn’t support Professor Stephen’s conclusions.
If you read that article in Bay Nature, you should also read this response to it, which explains that the statements made about embers are entirely inaccurate. The statements do not correspond to studies of actual wildfires by US Forest Service, including the 1991 fire, and they are also contradicted by the FACTS about that fire. For example, the fire covered only 3 miles at its widest point, which makes the statement that eucalyptus embers are capable of traveling 20 miles and arriving still on fire utterly ridiculous. If you know Zach, I suggest you speak with him. He has learned a lot since writing that article.
I would add to this discussion the fact that there were no houses in this area which would have added a LOT of fuel to the fire. (Unless properly constructed and maintained to be ignition resistant) Lessons learned: focus on robust fire response, carefully thought out evacuation plans, leave/plant trees, avoid ignitable construction. Thin the houses, not the trees.
This is ignoring the fact that a grass fire can be fought. Once a fire reaches the tree canopy and becomes a crown fire, all one can really do is watch it burn, especially in forests where canopies are fairly closed (oak woodlands generally have distance between trees that reduce the likelihood of this happening). As the leaves and branches high in the canopy burn and float through the wind, they create hot spots (especially near houses, when they fall on roofs or gutters). As Dr. Stephens as noted, eucalyptus are notorious for spreading embers in this fashion over vast distances. Minimizing the opportunity for a fire to crown is key. Logs on the ground are not ideal but they do not really facilitate crowning. Had that fire crowned, all the trees would have died. It is likely many people would have as well due to the close proximity to large office spaces. Oaks are also a long-lived species adapted to this ecosystem. Nearly all the Monterey pines in this are dead or dying before they reach 100 years old due to western gall rust. Eucalyptus are showing similar mortality issues (look at UCSF or the grove on the UC Campus) due to a butt rot disease that rots them from the inside. These trees will die, and research has shown they are not regenerating themselves, either. Should we be managing to make sure we have the best chance of having forest there in the future by promoting the naturally vegetated condition, which also provides the best balance of nice areas for recreation, wildlife habitat, and safety; or should we just leave it until the “big one” hits and the whole ecosystem is annihilated?
When tall dry grass ignites, it has higher flame lengths that burn more intensely than forest fires. The major fires in California, the ones that last for days, have been grassland and brush (chaparral) fires. The key to preventing crown fires is the removal of ladder fuels, the understory made up of dry weeds that could cause the fire to move up into the crown. It is highly unlikely that the Grizzly fire would have spread to “large office spaces,” since the response by fire fighters was excellent and fire fighters from several agencies worked together to stop the fire from spreading. The Grizzly fire was entirely different from the 1991 Oakland fire, which spread by means of one wooden house torching the next one.
The grim scenarios described by both Stephens and Roberts are entirely speculative AND they have nothing to do with reality. In fact, the fire did NOT spread into the tree canopy, nor did embers start spot fires. The 1991 fire was a wind driven fire that did ignite tree canopies, but they ignited ALL species of trees, including oaks and redwoods. Read Margaret Sullivan’s book, Firestorm, based on interviews with witnesses of that fire for confirmation of that FACT. The FEMA Technical report on the ’91 fire said embers that started spot fires were from “brush.” A US Forest Service study of embers cast by wildfires all over the world said that the only identifiable ember in the ’91 fire was a cedar shingle from one of the burned homes. Oaks do not live longer than eucalyptus. US Forest Service tree database says coast live oaks live about 200-250 years. Blue gum eucalyptus live in Australia from 300-500 years. They haven’t been here that long, so we don’t how long they will live here, but certified arborists with no nativist bias say they are healthy here and they expect them to live another 100-200 years. Quotes from the Presidio forester confirm that FACT. In contrast, many of our coast live oaks are being killed by Sudden Oak Death. A study published in April 2015, predicted that all coast live oaks in California would eventually be killed by SOD. Inform yourselves of the FACTS before spinning scary tales.

Add new comment