We used 5-year MODIS observations to analyse the effects of land conversion, from natural vegetation to agriculture, on the extent of areas burned in Africa. The results show that, in the vegetation units most affected by fire, the decrease in annual area burnt is related to the increase in agricultural land.

Satellite-derived measurements were used to compare prescribed fires and wildfires with respect to intensity and smoke plume area for the regions around the Australian cities of Melbourne and Sydney. Both fire types formed a continuum of intensity and plume size. We suggest that overall smoke plume area from the landscape is substantially smaller from prescribed than from wildfires.

Smoke plume height is important for understanding and simulating smoke effects on air quality. Plume height was measured using a ceilometer for prescribed burns in the south-eastern US. Average smoke plume height is ~1 km and trends upward from winter to summer, that being opposite to the seasonal trend in smoke intensity. Smoke plume height fluctuates at the time scales from minutes to hours for most burns.

A fast, simple, free-agent CA fire model is described and validated with an experimental grassfire, FireFlux. The model accurately simulated observed fire behaviour (spread rate, fire line depth and fire intensity) and bulk patterns of measured time-series of 2-m winds at two towers. Spatial windfields compared favourably with results from full-physics fire models.

Detailed measurements of convective and radiative heating rates in wildland fires are presented. Results indicate that the relative contribution of each to total energy release is dependent on fuel and environment.

The effect of sampling rate on interpretation of peak and average heating in wildland flames were explored. Results indicate that sampling rate is critical when considering temporally dependent response but less critical to average or cumulative response.

We compared three wildfire ignition distribution models for a national forest in Michigan, USA, in terms of their predictive accuracies, variable rankings and prediction maps. Although predictive accuracies and variable rankings were similar among models, prediction maps varied considerably, highlighting the potential need to compare multiple models for wildfire ignitions.

Frequently burned chamise chaparral shrublands in San Diego County, California, were mapped and analysed to investigate changes in vegetation condition and composition associated with short intervals between fires. Results indicate that chamise chaparral is vulnerable to alteration and type conversion when fire return intervals are 4–5 years or less.

Although many studies have reconstructed fire history in different forest types of Patagonia, little is known about fire regime in the Argentinean Araucaria araucana forests. In this study, fire history in ten sites was reconstructed and fire frequency was analysed in relation to human activities and major climate forcing.

A fire history of northern larch forests in Central Siberia was studied. The average fire return interval was found to be 295 ± 57 years, decreasing with latitude decrease. Post-Little Ice Age warming caused an approximate doubling of the fire frequency.

A stated preference survey of US federal wildfire managers was used to study the tradeoffs managers are willing to make among a variety of fire-suppression objectives, including protection of homes and a highly valued watershed, personnel safety, fire duration and suppression cost.

In this manuscript, we explore the availability and sufficiency of extant USA federal land management agency aviation data to track airtanker use and cost trends. Recent aviation and incident records indicate that the use of airtankers tends to occur on extended attack and large-fire support rather than the intended primary use of initial attack. Improved data collection on flight objective and drop locations is recommended.

Results from a stated preference survey indicate that most homeowners in Florida living in fire-prone communities tend to under-invest in programs designed to reduce wildfire risk. Willingness to pay for wildfire protection programs increases for homeowners who have prior personal experience with the consequences of wildfire. Therefore, these people may be influential in encouraging other community members to invest in wildfire protection.

Fireline production rates are used in decision making for fire suppression activities. Both the under and over utilisation of fire suppression inputs can have substantial consequences for the cost of fire suppression. This paper finds that standard fireline production rates are much greater than the rates estimated using economic production models.

Given the goal of protecting homes from wildfire, eliminating flammable vegetation around all homes in the western US (which would reduce fire risk) would require treating a much smaller area than if beetle-affected forests (>98% of which are in remote areas) were prioritised (which would not reduce fire risk).