SCIENTIFIC PROGRAMS AND ACTIVITIES
|August 21, 2014|
This 2-day workshop will focus on the mathematical and statistical modelling of the spread of biological and physical processes in forests. Individuals who have studied insect infestations in forests will be invited as well as researchers into the mechanisms underlying the spread of forest fires. In particular, we will learn from specialists in Mountain Pine Beetle as well as Spruce Budworm.
In the past two years, there has been a wave of inter-disciplinary
research activity involving mathematicians and statisticians
The present workshop will build on the earlier successes. While the earlier workshops were fairly broadly based in forest ecology and forest fire science, our goal for this workshop is to have a somewhat more narrow focus: spread modelling -- of fire and of pests.
The forest fire research community has made steady progress over the past four decades, in increasing our understanding of the nature of forest fires. Mathematical models for predicting fire occurrence have been developed. Deterministic spread models are being implemented for planning purposes and for use in computer simulations to aid in prediction of the future behavior of existing and potential fires. Such models are used in conjunction with queueing models in the strategic management of fire-fighting resources such as aircraft and fire fighters. Although much has been learned about the interactions of weather and fuel-types and their effects on fire spread, and intensity a large number of questions remain. For example, how can jump-fires ignited by burning bark and other firebrands carried by the wind, in advance of a spreading fire, be modelled as a stochastic process? How can the fire hazard in particular areas be estimated reliably? What are the potential impacts of climate change on fire regimes and fire management systems?
We plan to have researchers involved with the Prometheus Fire Spread
model at this workshop. There are several mathematical issues connected
with this model and its implementation which need to be addressed.
A level set approach has been suggested as an alternative, and details
on this alternative will be considered at this workshop. Incorporating
randomness into Prometheus is another major goal; the possibilities
will be laid out at this workshop. It should be noted that Prometheus
was featured as one
Other stochastic models for fire spread will also discussed, including interacting particle systems. A session on wind modelling is also envisioned. A director of Western's Boundary Layer Wind Tunnel will be present to discuss models for windborne debris. Ecological effects of these kinds of disturbances will also be considered. Interactions between fire and ecology will be discussed.
Models of the spread of insects through forests will also be considered. Here, spatial statisticians play a major role. In particular, Markovian mixed effects models have been shown to be an effective way of modelling outbreaks such as that of the mountain pine beetle and spruce budworm.
The format of the workshop will be similar to the BIRS workshop held in 2006; there will be some talks, but there will also be generous amounts of time set aside for roundtable discussion as well as informal discussion. We want to encourage the math/stat researchers to interact directly with their forest research counterparts.
A block of standard rooms are available at Windermere Manor (very near campus) at $106 per night. These rooms will be released on October 20 so book soon.