Risk analysis evolved out of the need to make decisions concerning highly stochastic events, and is well suited to analyse the likelihood, intensity and potential effects of wildfires. This paper reviews recent advances in fire risk analysis in fire prone systems around the world.

We reviewed over 200 articles in the wildland fire social science literature to identify key lessons in the following topic areas: social acceptability of fire and fuels management, risk perception, homeowner and community preparedness, citizen–agency interactions, trust, communication and outreach, public responses during and after a wildfire, recreation, planning and policy and organisational effectiveness.

The paper reviews some physical processes involved in wildland fire spread. Commonly made assumptions of how radiation and convective heat transfer lead to ignition of live and dead fuel particles are not supported experimentally. An experimentally supported theory of ignition is the needed basis for models of wildland fire spread.

Moving from wildfire suppression to broad fire management requires a risk-analysis approach and integrated developments in core fire science and decision-support technology. Current gaps in science, data and technology hinder risk-based management. We summarise science gaps and discuss a common analytic risk-based framework.

A synthesis of approaches for describing wildland surface fuel loadings illustrates major problems with classifying fuels for fire management. The next generation of fuel description systems must account for the spatial and temporal variability of fuel properties.

The 12-site Fire and Fire Surrogate study evaluated ecological effects of different treatments – prescribed fire and mechanical thinning – designed to reduce excessive fuel in dry USA forests. Treatments made stands more able to withstand moderate wildfire. Short-term effects on soils, plants and animals were subtle and differed for fire v. mechanical treatments.

Among the key issues in smoke management is predicting the magnitude and location of smoke effects. This review follows the development of these tools from various indices and simple screening models to complex air quality models with a focus on how each tool represents key processes involved in smoke transport.

Several tools have been developed to support post-wildfire assessment and treatment decision making, such as prediction models, field measurement methods, research syntheses, reference catalogues and databases, and calculation spreadsheets for resource valuation. These tools provide relevant science in usable formats to facilitate the use of current knowledge in post-wildfire management decisions.