Martin E. Alexander

Assessing crown fire potential in coniferous forests of western North America: a critique of current approaches and recent simulation studies

To control and use wildland fires safely and effectively depends on creditable assessments of fire potential, including the propensity for crowning in conifer forests. Simulation studies that use certain fire modelling systems (i.e. NEXUS, FlamMap, FARSITE, FFE-FVS (Fire and Fuels Extension to the Forest Vegetation Simulator), Fuel Manage- ment Analyst (FMAPlus), BehavePlus) based on separate implementations or direct integration of Rothermels surface andcrownrateoffirespread modelswithVanWagnerscrownfiretransitionandpropagationmodelsareshowntohavea significant underprediction bias when used in assessing potential crown fire behaviour in conifer forests of western North America.Theprincipal sourcesof thisunderprediction biasareshowntoinclude:(i)incompatible modellinkages;(ii)use of surfaceand crownfire rateof spreadmodels that havean inherent underprediction bias;and (iii)reductionincrown fire rate of spread based on the use of unsubstantiated crown fraction burned functions. The use of uncalibrated custom fuel models to represent surface fuelbeds is a fourth potential source of bias. These sources are described and documented in detail based on comparisons with experimental fire and wildfire observations and on separate analyses of model components.Themannerinwhichthetwoprimarycanopyfuelinputsinfluencingcrownfireinitiation(i.e.foliarmoisture contentandcanopybaseheight)ishandledinthesesimulationstudiesandthemeaningofScottandReinhardtstwocrown fire hazard indices are also critically examined.

Science, technology, and human factors in fire danger rating: the Canadian experience.

The present paper reviews the development of the Canadian Forest Fire Danger Rating System (CFFDRS) and its implementation in Canada and elsewhere, and suggests how this experience can be applied in developing fire danger rating systems in other forest or wildland environments. Experience with the CFFDRS suggests that four key scientific, technological, and human elements need to be developed and integrated in a national forest fire danger rating system. First among these is a sustained program of scientific research to develop a system based on relationships between fire weather, fuels, and topography, and fire occurrence, behavior, and impact appropriate to the fire environment. Development of a reliable technical infrastructure to gather, process, and archive fire weather data and to disseminate fire weather forecasts, fire danger information, and fire behavior predictions within operational agencies is also important. Technology transfer and training in the use of fire danger information in fire operations are necessary, as are cooperation and communication between fire management agencies to share resources and set common standards for information, resources, and training. These elements must be appropriate to the needs and capabilities of fire managers, and must evolve as fire management objectives change. Fire danger systems are a form of media; system developers should be careful not to overemphasize scientific and technological elements at the expense of human and institutional factors. Effective fire danger systems are readily assimilated by and influence the organizational culture, which in turn influences the development of new technologies. Most importantly, common vision and a sense of common cause among fire scientists and fire managers are needed for successful implementation of a fire danger rating system.

Interdependencies between flame length and fireline intensity in predicting crown fire initiation and crown scorch height

This state-of-knowledge review examines some of the underlying assumptions and limitations associated withtheinter-relationships amongfourwidelyuseddescriptorsofsurfacefirebehaviourandpost-fireimpacts inwildland fire science and management, namely Byrams fireline intensity, flame length, stem-bark char height and crown scorch height. More specifically, the following topical areas are critically examined based on a comprehensive review of the pertinent literature:(i)estimatingfirelineintensityfrom flamelength;(ii)substitutingflamelengthforfirelineintensityin Van Wagners crown fire initiation model; (iii) the validity of linkages between the Rothermel surface fire behaviour and Van Wagners crown scorch height models; (iv) estimating flame height from post-fire observations of stem-bark char height; and (v) estimating fireline intensity from post-fire observations of crown scorch height. There has been an overwhelming tendency within the wildland fire community to regard Byrams flame length-fireline intensity and Van Wagners crown scorch height-fireline intensity models as universal in nature. However, research has subsequently shown that such linkages among fire behaviour and post-fire impact characteristics are in fact strongly influenced by fuelbed structure, thereby necessitating consideration of fuel complex specific-type models of such relationships. Additional keywords: fire behaviour, fire impacts, fire modelling, first-order fire effects, flame angle, flame depth, flame-front residence time, ignition pattern, stem-bark char height, surface fire.

Uncertainty associated with model predictions of surface and crown fire rates of spread

The degree of accuracy in model predictions of rate of spread in wildland fires is dependent on the models applicability to a given situation, the validity of the models relationships, and the reliability of the model input data. On the basis of a compilation of 49 fire spread model evaluation datasets involving 1278 observations in seven different fuel type groups, the limits on the predictability of current operational models are examined. Only 3% of the predictions (i.e. 35 out of 1278) were considered to be exact predictions according to the criteria used in this study. Mean percent error varied between 20 and 310% and was homogeneous across fuel type groups. Slightly more than half of the evaluation datasets had mean errors between 51 and 75%. Under-prediction bias was prevalent in 75% of the 49 datasets analysed. A case is made for suggesting that a ?35% error interval (i.e. approximately one standard deviation) would constitute a reasonable standard for model performance in predicting a wildland fires forward or heading rate of spread. We also found that empirical-based fire behaviour models developed from a solid foundation of field observations and well accepted functional forms adequately predicted rates of fire spread far outside of the bounds of the original dataset used in their development. We examined error statistics of operational wildland fire spread models.We compiled 49 fire spread model evaluation datasets involving 1278 observations.Mean percent error varied between 20 and 310% and was homogeneous across fuel type groups.The analysis suggests that a ?35% error interval is a reasonable standard for model adequacy.

Short communication: Are the applications of wildland fire behaviour models getting ahead of their evaluation again?

Evaluation is a crucial component for model credibility and acceptance by researchers and resource managers. The nature and characteristics of free-burning wildland fires pose challenges to acquiring the kind of quality data necessary for adequate fire behaviour model evaluation. As a result, in some circles it has led to a research culture that tends to avoid evaluating model performance. Operational fire modelling systems commonly used in western North America have been shown to exhibit an underprediction bias when employed to determine the threshold conditions necessary for the onset of crowning and the associated spread rate of active crown fires in conifer forest stands. This pronouncement was made a few years ago after at least a decade of model misapplication in fire and fuel management simulation modelling stemming from a lack of model evaluation. There are signs that the same situation may be repeated with developing physics-based models that simulate potential wildland fire behaviour; these models have as yet undergone limited testing against observations garnered from planned and/or accidental wildland fires. We propose a broad co-operative project encompassing modellers and experimentalists is needed to define and acquire the benchmark fire behaviour data required for model calibration and evaluation.

Assessing the effect of foliar moisture on the spread rate of crown fires

Thispaperconstitutesadigestandcritiqueofthecurrentlyavailableinformationpertainingtotheinfluenceof live fuel or foliar moisture content (FMC) on the spread rate of crown fires in conifer forests and shrublands. We review and discuss the findings from laboratory experiments and field-based fire behaviour studies. Laboratory experimentation with single needles or leaves and small conifer trees has shown an unequivocal effect of FMC on flammability metrics. AmuchlessdiscernibleeffectofFMConcrownfirerateofspreadwasfoundintheexistingsetofexperimentalcrownfires carried out in conifer forests and similarly with the far more robust database of experimental fires conducted in shrubland fuel complexes. The high convective and radiant heat fluxes associated with these fires and the lack of appropriate experimental design may have served to mask any effect of FMC or live fuel moisture on the resulting spread rate. Four theoreticalfunctionsandoneempiricalfunctionusedtoadjustrateoffirespreadfortheeffectoffoliarorlivefuelmoisture were also concurrently examined for their validity over a wide range of FMC conditions with varying outcomes and relevancy. None of these model functions was found suitable for use with respect to dead canopy foliage. Additional keywords: conifer forest, crowning, dead foliage, fire behaviour, flammability, foliar moisture content, fuel moisture, heat transfer, live fuel moisture, shrubland.

A mathematical model for predicting the maximum potential spotting distance from a crown fire

A mathematical model is presented for predicting the maximum potential spot fire distance from an active crown fire. This distance can be estimated from the height of the flame above the canopy top, wind speed at canopy-top height and final firebrand size (i.e. its residual size on alighting), represented by the diameter of a cylinder of woody char. Thecompletemodelsystemcomprisesseveralsubmodelsorcomponents:amodelfortheheightandtiltangleofthewind- blown line-fire flame front, a simplified two-dimensional model of the wind-blown buoyant plume from the fire, an assumed logarithmic wind speed variation with height, and an empirically based model for the burning rate of a wooden cylinder incross flow,which represents the firebrand. The trajectory of the burning particle is expressed analytically from where it leaves the lower boundary of the plume until it enters the canopy top. Adding the horizontal distance of this flight to that of the point where the particle can no longer be held aloft by the plume flow gives a spotting range that depends on the final diameter of the burning particle. Comparisons of model output with existing information on crown fire spotting distances has initially proved encouraging but further evaluation is warranted. Additional keywords: canopy-top height, crowning, ember, extreme fire behaviour, fire plume, firebrand, flame height, ignition, spot fire, wind speed.

Empirical-based models for predicting head-fire rate of spread in Australian fuel types

Summary The knowledge of a free-burning fire’s potential rate of spread is critical for safe and effective bushfire control and use. A number of models for predicting the head-fire rate of spread in various types of Australian vegetation have been developed over the past 60 years or so since Alan G. McArthur began his pioneering research into bushfire behaviour. Most of the major Australian vegetation types have had more than one model developed for operational use. These include grassland, shrubland, both dry and wet eucalypt forests, and pine plantation fuel types. A better understanding of the technical basis for each of these models and their utility is essential for the correct selection and application of the most appropriate models. This review provides a systematic overview of 22 models of the rate of fire spread and their applicability in prescribed burning and wildfire operations. Background information and a description of each model is given. This includes information on the data used in the model development that defines the bounds of its application. The mathematical equations that represent each model are given along with a discussion of model form and behaviour, the main input variables and their influence, and evaluations of model performance undertaken to date. This review has enabled the identification of those models that constitute the current state of knowledge with respect to bushfire behaviour science in Australia. We recommend the models that should underpin best practice in the near term in the operational prediction of fire behaviour and those that should no longer be used, and provide reasons for these recommendations.

Fire danger monitoring using RADARSAT-1 over northern boreal forests

The goal of this study was to evaluate the potential use of RADARSAT‐1 images to assess daily variations in dead fuel moisture over a northern boreal forest area, as parameterized by the Canadian Forest Fire Weather Index (FWI) System. The study area was located in the south‐central region of Canadas Northwest Territories and was comprised of three land cover classes (mature forest, burned forest, and fireguard areas). Nineteen RADARSAT‐1 images were acquired over the study area in June 2000 and August 2002, and their backscatter was compared to weather data and to the FWI System components. In both cases, the influences of incident angle and land cover class were measured. Radar backscatter was related to rainfall, and strong relationships were observed with the FWI System codes and indexes, particularly for Duff Moisture Code (R 2 = 0.68–0.83), Drought Code (R 2 = 0.77–0.82), Build‐up Index (R 2 = 0.72–0.86), and FWI (R 2 = 0.62–0.85). The best regression models were obtained using a stepwise regression procedure in which radar backscatter from the burned forest was used as the independent variable.

Modelling the effects of surface and crown fire behaviour on serotinous cone opening in jack pine and lodgepole pine forests

A methodology has been developed for defining the various threshold conditions required for the opening of serotinous cones and viable seed release in the overstorey canopies in jack pine (Pinus banksiana) and lodgepole pine (Pinuscontortavar.latifolia)forestsonthebasisoffirelineintensityand,inturn, rateoffirespreadandfuelconsumption. The extent of the effects to the overstorey canopy (i.e. crown scorch height and flame defoliation) and the type of fire (i.e. low- to high-intensity surface, intermittent crown and active crown) vary at any given fireline intensity level and are principallyafunctionoffoliarmoisturecontent,canopybaseheight,standheightandcanopybulkdensity.Theviabilityof the seed stored in serotinous cones of the two pine species begins to decreases once the flame-front residence time at the ground level of an active crown fire exceeds 50s. Additional keywords: convection column temperature, crown scorch height, fire ecology, fireline intensity, flame front residence time, fuel consumption, rate of fire spread, tree regeneration.