Wildfire Modeling and Prevention
A wildfire prevention plan starts with modeling where wildfire risk exists, then proceeds into developing mitigation strategies, short term and long term, for individual "firesheds" (areas defined by topography and the nature of wildfire).
In 2022, Dr. John Radke, in conjunction with UC's Center for Catastrophic Risk Management (CCRM), developed a proposal for Orinda to model wildfire risk throughout Orinda, at a detail that brought it down to individual properties.
The primary tool that Radke uses to predict fire intensity is called FlamMap. It was first developed in 2006 to predict the behavior of forest fires. Predicting fire behavior is complex. In addition to fuel load several other factors need to be factored in including terrain and weather (wind, humidity and temperature).
What the tool produces is a map of the intensity of the fire at any point. In his proposal to the City, Radke provided an example for one area in Orinda (between Scenic Dr. and Estates Dr.), shown below.
The image measures projected flame length (for a given weather assumption). Flame length, in turn, reflects the intensity of the fire, which can be interpreted via the "Hauling Chart" (below).
Flame lengths in excess of 12 feet represent energy levels in excess of 1,000 BTU per square foot which is beyond what firefighters can get in and fight. In other words, this "fireshed" would burn out of control and every home in it would probably be destroyed. To prevent this, vegetative fuel must be removed to the degree that the fire does not exceed energy levels preventing suppression efforts.
The model can also project the time it takes for a wildfire to progress over the terrain, again with weather assumptions, mainly wind speed.
This is one out of 100 firesheds that Radke identified in Orinda.
The FlamMap model was originally designed to measure fire intensity and spread in a forest environment. While the analytics are complex, there is one "simplifying" factor in a forest and that is the fuel load is relatively homogeneous. This is not the case in Orinda. There is a huge variety of vegetation ranging from relatively "innocuous" oaks and redwoods to extremely volatile eucalyptus, pines and junipers. And the existence of ground vegetation (what are called "ladder fuels") is critical.
There is nothing "homogeneous" about the vegetation in Orinda. This is why Radke's proposal included two elements that may be considered "theory", but which are actually "fine tuning" for more accurate projections. They both increase the level of detail which the model uses to predict.
The first is the area of each data "point". In a forest, a square ten meters (33 feet) on a side is sufficient due to the homogeneity of foliage. Radke planned to use much more precise data, a one-meter square, 100 times the accuracy. With each data point having its own fuel load designation.
The second element is identifying the vegetation in each data point. Some of this can be derived from satellite and "lidar" data, but Radke also proposed using a program called SpectoBase (page 14 of his proposal) to allow a detailed inventory of vegetation on a property.
Members of the SSTOC (Supplemental Sales Tax Oversight Commission - overseeing the spending of Measure R sales tax funds) which Radke met with three times, saw the use of this tool as an opportunity to educate the community regarding wildfire. They envisioned school kids taking the SpectoBase application and cataloging their properties and their neighbors' properties.
While it would be nice to get SpectoBase data for all 7,000 properties in Orinda, another new tool which could be used to "fill in the gaps". That would be Artificial Intelligence (AI). Using known data to "fill in the gaps" for areas that have not been detailed. It would not be a perfect solution, but it would increase the accuracy of the projections. Again, this may be "research" but very useful research which could make Orinda safer. (Radke and his sons, one of which has a PhD in AI, published a paper in a leading academic journal on the subject.)
Another element of "fuel load" which the model does not take into account is that of the homes themselves. They are obviously a huge fuel source and if a fire goes unchecked, they multiply the intensity. The goal would be to keep the fire intensity low enough to prevent massive home involvement. Not that every home would remain unscathed, but if the intensity was kept low enough, firefighters can respond and suppress structure fires.
Once the areas of risk are defined, a mitigation strategy for each neighborhood (parcel by parcel) would need to be designed. This could include action by the property owner, especially adjacent to the home, but it would also include "common" areas over which the "government", with permission from the property owners, would take responsibility.
In the above "Figure 10", the homes are on the periphery of the fireshed. The common area, the area of greatest flame intensity, could be mitigated, and even maintained, by the City and MOFD. Recently, a vegetation mitigation expert evaluated this fireshed. Even though the City and MOFD have been actively engaging property owners to mitigate their own properties for several years, and this particular area was highlighted by Radke as highly at-risk two years ago, it was estimated that it would cost $71,000 to appropriately mitigate the common area and $15,000 a year to maintain it. This is one of 100 separate firesheds in Orinda.
Two other firesheds identified by Radke were also inspected by the vegetation mitigation expert. One ran from Valley View Drive up to Crestview Drive, and the other was in the area surrounding Tiger Tail Ct. It was estimated that the Valley View fireshed would cost about $83,000 to mitigate and $17,000 a year to maintain while Tiger Tail would cost $192,000 to mitigate and $31,000 a year to maintain.
Total cost for all three firesheds: $346,000 for mitigation and $63,000 a year to maintain. Assuming these three are "representative" firesheds, extrapolating to all of Orinda's 100 firesheds would bring the cost to maximize the safety of Orinda being $11.5 million and $2.1 million a year to maintain.
This is well within the means of Orinda and MOFD to finance. Orinda's Measure R sales tax is generating almost $4 million a year and Orinda's property taxes going to MOFD are $4.5 million in excess of what MOFD spends to currently service Orinda. Plus, MOFD currently has $28 million in reserves and projects that to increase to $62 million in ten years.