bushfire projects (Prof John Dold)

please contact Prof Dold for further information

There are several other project topics. To give a flavour, two possible topics are:


Bushfires or wildfires are of great importance ecologically, affecting habitat, flora and climate, as well as economically. Fires last year in Greece and California caused losses in excess of US$3bn. With global warming they are showing signs of increasing in frequency and severity, while also probably making the problem worse. They can become very hazardous to life and property if they are not controlled.

Along with collaborators in Manchester and overseas, Prof Dold is studying many aspects of wildfire behaviour.

A bushfire normally spreads as a fireline. The flames above the fireline cause fresh vegetation to be heated, mainly by radiation, until it ignites and so moves the fireline forwards. The way in which this happens is strongly influenced by wind and slope.

Some background information on bushfires can be seen via this link and a lecture by Prof Dold on some basic bushfire modelling can be found via this link.


effects of air-flow and entrainment around bushfires

One project will focus on the air-flow around a fireline and the way in which this is influenced by the shape and intensity of the fireline. The influence also goes the other way because the rate of spread and the intensity of the fireline are influenced by the air-flow. This project aims to provide a sound physical and mathematical basis for describing data from fires in Australia that have not yet been adequately explained.

A foundation for the study has already been laid with plans in place for further experimental and theoretical work involving air flows induced by firelines and the spread-rate of the fireline in response to the air flow.

The project will examine the fireline as a moving interface, spreading at a rate that depends in a prescribed way on the air-flow into the fireline. The flow of ambient air (as opposed to flow in the hot buoyant plume above the fire) can be approximated by treating the fire as sucking in air; the entrained air is heated and ejected upwards in the plume.

The overall air flow combines the wind and the effects of suction over the entire fireline, which can be solved numerically as a boundary integral for flow in an ideal fluid. Data is available for the spreading of fairly large experimental grassfires in Australia.

The first objective of the project will be to solve for moving firelines as initial value problems over two dimensional surfaces under given wind conditions, with fire-induced air-flow influencing the spread of the fire. The second will be to find suitable model parameters, through numerous targeted numerical calculations, that match up with the experimental fires.

(contact Prof Dold for further information)


ember storms

Another way in which a fire can sometimes spread is through the effect of burning embers carried by the air. If these are still burning as they land in unburnt vegetation they can start a new fire. In fact embers are considered to be the main mechanism by which houses are set alight as bushfires enter built-up areas.

In very windy conditions embers can completely dominate in the overall spreading of a bushfire, creating what may be called an "ember storm". In such a situation, embers ignite many small fires that provide sources for new embers to be carried by the wind. Each small fire can spread in the normal way but, in the overall picture, wind-borne embers are the main factor that drives the bushfire forwards.

The project is aimed at developing a discrete simulation model for ember storms by combining two aspects. One is the simulation of the spreading of many small fires as they grow and coalesce over a two-dimensional surface; the level-set method could be used for this, although other approaches may prove equally effective. The other aspect is the role of embers.

Embers can be taken to be lofted randomly with a random size distribution at a prescribed probability that depends on the intensity of burning and the wind-speed. The distance carried by the wind and the chance of the ember sill being alight when it lands can then be modelled, and included in the fire-spread simulations by their initiation of new fires.

The main objective is to gain an understanding of the way in which the overall spreading of an ember storm depends on the assumptions made about the way in which embers are generated and transported. This will help in gaining a better understanding of the factors that control them and will link into other possible ways of describing them mathematically.

(contact Prof Dold for further information)