David L. Martell

Forest management challenges for operational researchers

This is an overview of operational research work that has been carried out to support strategic forest management planning, short term forest planning, forest operations, and forest fire management. It identifies important unresolved problems and new challenges that could serve as a rich source of problems for both practitioners and researchers.

CLIMATE CHANGE AND PEOPLE-CAUSED FOREST FIRE OCCURRENCE IN ONTARIO

Climate change that results from increasing levels of greenhouse gases in the atmosphere has the potential to increase temperature and alter rainfall patterns across the boreal forest region of Canada. Daily output from the Canadian Climate Centre coupled general circulation model (GCM) and the Hadley Centres HadCM3 GCM provided simulated historic climate data and future climate scenarios for the forested area of the province of Ontario, Canada. These models project that in climates of increased greenhouse gases and aerosols, surface air temperatures will increase while seasonal precipitation amounts will remain relatively constant or increase slightly during the forest fire season. These projected changes in weather conditions are used to predict changes in the moisture content of forest fuel, which influences the incidence of people-caused forest fires. Poisson regression analysis methods are used to develop predictive models for the daily number of fires occurring in each of the ecoregions across the forest fire management region of Ontario. This people-caused fire prediction model, combined with GCM data, predicts the total number of people-caused fires in Ontario could increase by approximately 18% by 2020–2040 and50% by the end of the 21st century.

Spatial patterns of lightning-caused forest fires in Ontario, 1976-1998

The spatial pattern of forest fire locations is of interest for fire occurrence prediction and for understanding the role of fire in landscape processes. A spatial statistical analysis of lightning-caused fires in the province of Ontario, between 1976 and 1998, was carried out to investigate the spatial pattern of fires, the way they depart from randomness, and the scales at which spatial correlation occurs. Fire locations were found to be spatially clustered. Kernel estimation of the spatial pattern of lightning strikes on days when the dryness of the forest floor exceeded a designated threshold yielded clusters in the same areas as the lightning fire clusters.

Fire weather index system components for large fires in the Canadian boreal forest

Canadian Fire Weather Index (FWI) System components and head fire intensities were calculated for fires greater than 2 km 2 in size for the boreal and taiga ecozones of Canada from 1959 to 1999. The highest noon- hour values were analysed that occurred during the first 21 days of each of 9333 fires. Depending on ecozone, the means of the FWI System parameters ranged from: fine fuel moisture code (FFMC), 90 to 92 (82 to 96 for individual fires); duff moisture code (DMC), 38 to 78 (10 to 140 for individual fires); drought code (DC), 210 to 372 (50 to 600 for individual fires); and fire weather index, 20 to 33 (5 to 60 for individual fires). Fine fuel moisture code decreased, DMC had a mid-season peak, and DC increased through the fire season. Mean head fire intensities ranged from 10 to 28 MW m −1 in the boreal spruce fuel type, showing that most large fires exhibit crown fire behaviour. Intensities of individual fires can exceed 60 MW m −1 . Most FWI System parameters did not show trends over the 41-year period because of large inter-annual variability. A changing climate is expected to create future weather conditions more conducive to fire throughout much of Canada but clear changes have not yet occurred.

A model for predicting human-caused wildfire occurrence in the region of Madrid, Spain

This paper describes the development and validation of a spatio-temporal model for human-caused wildfire occurrence prediction at a regional scale. The study area is the 8028-km2 region of Madrid, located in central Spain, where more than 90% of wildfires are caused by humans. We construct a logistic generalised additive model to estimate daily fire ignition risk at a 1-km2 grid spatial resolution. Spatially referenced socioeconomic and weather variables appear as covariates in the model. Spatial and temporal effects are also included. The variables in the model were selected using an iterative approach, which we describe. We use the model to predict the expected number of fires in our study area during the 2002–05 period, by aggregating the estimated probabilities over space–time scales of interest. The estimated partial effects of the presence of railways, roads, and wildland–urban interface in forest areas were highly significant, as were the observed daily maximum temperature and precipitation.

Forest Fire Management

Publisher Summary Fire management programs typically include prevention measures to reduce the number of people-caused fires that occur, detection systems to find fires while they are small, initial attack systems to contain fires before they burn over large areas, and large fire management systems that are designed to minimize the damage that results from large fires that are not controlled by the initial attack system. They also include fuel modification measures to mitigate the impact of fires that do occur and the use of prescribed fire to fulfill silviculture, wildlife habitat management, and other land management objectives. Fire management also calls for conscious decisions to allow some wildfires to burn freely or to be subjected to limited suppression action if and when the net benefit of doing so is thought to be positive. Societies that are confronted with potentially destructive forest or wildland fires develop fire management organizations to modify fires impact on people, the things they value, and the ecosystems about which they are concerned. The social, economic, and political institutions that control it determine a fire management organizations objectives.

Carbon Release from Fires in the North American Boreal Forest

Using current estimates of exchanges of carbon between the atmosphere, ocean, and biosphere, researchers have determined that a movement of as much as 2 Gt C yr-1 out of the atmosphere pool is unaccounted for (Schlesinger 1991). It has been hypothesized that this missing carbon sink may be found in the terrestrial biosphere and most likely in northern latitudes (Tans et al. 1990; Ciais et al. 1995; Fan et al. 1998; Kasischke, this volume, Chapter 2). Research is underway to model and quantify the carbon cycle in northern (temperate and boreal) forest ecosystems a biogeochemical cycle that is poorly understood and very difficult to quantify because of the high degree of spatial and temporal variability and complexity in distribution of different forest types and ecosystem processes.

The impact of fire suppression, vegetation, and weather on the area burned by lightning-caused forest fires in Ontario

We describe the development of a statistical model of spatial variation in the area burned by lightning-caused forest fires across the province of Ontario. We partitioned Ontario’s fire region into 35 compartments, each of which is relatively homogeneous with respect to its vegetation, weather, and the level of fire protection it receives. We used linear regression and spatial autoregressive models to relate the average annual area burned in a compartment to its vegetation, weather, and level of protection attributes. We also examined the relationship between burned area and the level of protection in two areas that are relatively homogeneous with respect to vegetation and weather. We found a statistically significant relationship between the average annual fraction of the area of a compartment burned by lightning-caused forest fires and its vegetation, weather, and the level of fire suppression effort it receives.

Some potential carbon budget implications of fire management in the boreal forest

Forest fire is a natural force that has shaped boreal and temperate forest ecosystems in the northern hemisphere for millennia, to the point that these forests have not only adapted to, but are dependent upon, periodic stand-replacement wildfires to maintain their existence. Over the past century people throughout northern forest ecosystems have coexisted, at times somewhat uneasily, with this important natural force, as fire management agencies attempted to balance public safety concerns and the industrial and recreational use of these forests, with costs, and the need for natural forest cycling through forest fires. In recent years fire management has been complicated by a growing recognition of the need to reconcile attempts to minimise fire losses with the growing cost of fire management and the beneficial impacts of fire. Canadian, Russian, and American fire managers have always designated parts of the boreal zone, usually in northern regions, as “lower priority” zones that receive little or no fire protection, since fires that occurred there were thought to have little or no significant detrimental impact on public safety and forest values. The realisation that total fire exclusion is neither possible nor ecologically desirable, initiated a gradual move toward the widespread adoption of fire management strategies that prioritise protection of high-value resources while permitting burning in more remote areas. This is particularly true in the boreal forest regions of Canada, Russia, and Alaska where lower population densities and forest use allow more flexible fire management strategies.

Modelling the effect of spatial scale and correlated fire disturbances on forest age distribution

Abstract With the exponential model, Van Wagner (1978) gave us valuable insight in understanding stand age and forest age distribution in fire-disturbed landscapes. He showed that, under certain conditions, the probability distribution of the age of a stand subject to periodic renewal by fire is exponential. The extension of this model to the landscape-level results, also under certain conditions, in an exponential shape for the forest age distribution. Empirical studies have supported this hypothesis in some landscapes and not in others. The results are believed to depend on the size of the landscape in question, the patterns of fire disturbance, and changes in the disturbance regime over time and space. In this paper, we present additional insight into some of the fundamental factors that determine the forest age distribution. We analyzed some alternative spatial models of fire disturbance, and used a spatial simulation model (FLAP-X) to explore whether the forest age distribution has an exponential shape, and whether it would be stable or variable over time under different conditions. We use different spatial and temporal disturbance patterns, some of which represent correlation due to fire growth and episodes of high fire disturbance. We describe FLAP-X and give the results of computational tests based on hypothetical data. We found that, under characteristic boreal fire disturbance regimes, we should not expect to find forest age distribution stability even at very large spatial scales due to the spatial and temporal correlation of disturbances.