This study was conducted to increase understanding of possible roles and importance of local three-dimensionality in the forward spread of wildfire models. A suite of simulations was performed using a coupled atmosphere-fire model, HIGRAD/FIRETEC, consisting of different scenarios that varied in domain width and boundary condition implementation. A subset of the simulations was strictly two-dimensional in the streamwise and vertical directions, while another subset of simulations involved igniting a finite-length fireline. The remaining simulations were all three-dimensional and employed periodic boundary conditions in the cross-stream direction and a fireline spanning the entire cross-stream extent of the domain.The three-dimensional periodic simulations were compared with the two-dimensional simulations, and then briefly with the finite-length fireline simulations. The two-dimensional scenarios were constrained in their ability to represent inherently three-dimensional physical phenomena such as horizontal flow penetrating through the fireline between plumes of rising hot gas, and cross-stream heterogeneity in the windfield. Elimination of these three-dimensional flow patterns in two-dimensional simulations resulted in over prediction of spread rates in low velocity situations and under predicted spread rates in high wind speed scenarios. In the three-dimensional simulations, local cross-stream heterogeneities in temperature and velocities lead to penetration of hot gases through the fireline and onto unburned fuel. Three-dimensional fires presented a positive correlation between increasing ambient wind speed and rate of spread. Further investigation of finite length fires is required in order to understand the ramifications of fireline curvature. Published by Elsevier B.V.

Field experiments are one way to develop or validate wildland fire-behavior models. It is important to consider the implications of assumptions relating to the locality of measurements with respect to the fire, the temporal frequency of the measured data, and the changes to local winds that might be caused by the experimental configuration. Twenty FIRETEC simulations of International Crown Fire Modeling Experiment (ICFME) plot 1 and plot 6 fires were performed using horizontally homogenized fuels. These simulations enable exploration of the sensitivity of model results to specific aspects of the interpretation and use of the locally measured wind data from this experiment. By shifting ignition times with respect to dynamic measured tower wind data by up to 2 min, FIRETEC simulations are used to examine possible ramifications of treating the measured tower winds as if they were precisely the same as those present at the location of the fire, as well as possible implications of temporal averaging of winds or undersampling. Model results suggest that careful consideration should be paid to the relative time scales of the wind fluctuations, duration of the fires, and data collection rates when using experimentally derived winds as inputs for fire models.

Firebrand transport is studied for disc and cylindrical firebrands by modelling their trajectories with a coupled-physics fire model, HIGRAD/FIRETEC. Through HIGRAD/FIRETEC simulations, the size of possible firebrands and travelled distances are analysed to assess spot ignition hazard. Trajectories modelled with and without the assumption that the firebrands’ relative velocities always equal their terminal velocities are. Various models for the flight and combustion of disc- and cylindrical-shaped firebrands are evaluated. Eight simulations are performed with surface fuel fires and four simulations are performed with combined surface and canopy fuels. Firebrand trajectories without terminal velocity are larger than those from models with terminal velocity. Discs travel further than cylinders, as discs are aerodynamically more favourable. Thin discs burning on their faces and tall cylinders burning around their circumference have shorter lifetimes than thin discs burning from their circumference or longer cylinders burning from their ends. Firebrands from canopy fires, with larger size and potential to ignite recipient fuel, travel further than firebrands from surface fires. In the simulations, which included a line fire ignition in homogeneous fuels on flat terrain, the firebrand launching patterns are very heterogeneous, and the trajectories and landing patterns are dominated by the coupled fire–atmosphere behaviour.

Slope and fuel structure are commonly accepted as major factors affecting theway wildfires behave. However, it is possible that slope affects fire differently depending on the fuel bed. Six FIRETEC simulations using three different fuel beds on flat and upslope topography were used to examine this possibility. Fuel beds resembling grass, chaparral, and ponderosa pine forests were created in such a way that there were two specific locations with identical local fuel beds located around them. These fuel beds were each used for a flat-terrain simulation and an idealised-hill simulation in order to isolate the impacts of the topography without the complications of having different local fuels. In these simulations, fuel bed characteristics have a significant effect on the spread rate and perimeter shape of the fires on both flat ground and on the idealised smooth hill topography. The analysis showed that these simulated fires evolved as they travelled between the locations even on flat ground, and the accelerations and decelerations that affect the fire occurred at different times and at different rates depending on the fuel bed. The results of these simulations and analyses indicate that though some general principles are true for all fuel beds, there are differences in the way that fires react to non-homogeneous topographies depending on the fuel bed.

Fires caused by lightning or Native Americans were the major ecological factor in the borderlands region of Arizona, New Mexico, and Mexico prior to European settlement. Historical overgrazing and aggressive fire suppression have led to the encroachment of woody vegetation and accumulations of woody fuels in these grasslands. Ranchers associated with the Malpai Borderlands Group, state and federal land managers, and the staff of The Nature Conservancy agreed that re-introducing fire could improve landscape productivity and biological diversity. The ranching community was concerned that continued encroachment of woody plants would eventually affect its economic viability and result in the subdivision of ranches and the loss of its way of life. The parties in the borderlands group worked together to develop prescribed fire plans that have resulted in four landscape-level prescribed fires since 1995. They also developed a monitoring plan using established photo points, ground transects, aerial surveys, and remote sensing techniques to ascertain fire effects and to determine if modified procedures could be beneficial. The prescribed fires and results from the monitoring spurred research on wildlife habitat requirements, fire histories, fire behavior, the effects of cool or warm season burning on multiple resources, and interactions of livestock and wildlife grazing and fire. Information from prescribed fires such as those conducted by the borderlands group will become more important as anticipated changes in the region’s climate and vegetation produce new challenges to ranchers and agency land managers.

Fires caused by lightning or Native Americans were the major ecological factor in the borderlands region of Arizona, New Mexico, and Mexico prior to European settlement. Historical overgrazing and aggressive fire suppression have led to the encroachment of woody vegetation and accumulations of woody fuels in these grasslands. Ranchers associated with the Malpai Borderlands Group, state and federal land managers, and the staff of The Nature Conservancy agreed that re-introducing fire could improve landscape productivity and biological diversity. The ranching community was concerned that continued encroachment of woody plants would eventually affect its economic viability and result in the subdivision of ranches and the loss of its way of life. The parties in the borderlands group worked together to develop prescribed fire plans that have resulted in four landscape-level prescribed fires since 1995. They also developed a monitoring plan using established photo points, ground transects, aerial surveys, and remote sensing techniques to ascertain fire effects and to determine if modified procedures could be beneficial. The prescribed fires and results from the monitoring spurred research on wildlife habitat reFire Ecology Special Issue Vol. 5, No. 1, 2009 Gottfried et al.: Private-Public Collaboration to Reintroduce Fire Page 86

Ten simulations were performed with the HIGRAD/FIRETEC wildfire behaviour model in order to explore its utility in studying wildfire behaviour in inhomogeneous topography. The goal of these simulations is to explore the potential extent of the coupling between the fire, atmosphere, and topography. The ten simulations described in this paper include five different topographies, each run with two different ambient wind speeds of 6 and 12 m s–1. The five topologies explored are: an idealised hill (which serves as the base centerline for the other topographies), two variations of the hill with lateral gradients downwind from the ignition line (one sloping up from the ‘hill’ at the centerline to form an upward sloping canyon parallel to the ambient wind, and the other sloping down from the centerline to form a ridge parallel to the ambient flow), one with a second hill upwind of the ignition line such that the fire is ignited in the bottom of a canyon that runs perpendicular to the ambient wind, and finally a flat terrain. The four non-trivial topographies have the same profile along the centerline downwind of the ignition line to help assess the impacts of topographic gradients that are perpendicular to the ambient wind. It is hoped that analysis of these simulations will help reveal where point-functional models are sufficient, where topographically modified wind fields are needed, and where fully coupled fire and transport models are necessary to properly describe wildfire behaviour.

Species composition of Pinus ponderosa P.&C. Lawson var. scopulorum Engelm. forests in the sky islands of the Davis Mountains was evaluated and classified using ordination and classification techniques. Twelve study sites were established within Pinus ponderosa forests in the summers of 2002 and 2003. Overstory (tree) and understory (shrub and herbaceous) vegetation strata were sampled. Thirteen associations were recognized by combining four overstory groups and eight understory groups. The associations reflected a wide range of ecological conditions with Juniperus deppeana associations occupying dry sites and Pinus ponderosa associations occupying wetter sites. The major environmental factors responsible for the differences in vegetation of these forests were altitude, slope position, slope inclination, and soil moisture.

Seedfall, natural regeneration establishment, and growth of planted seedlings was observed from 1981 to 2001 under shelterwood and seedtree overstories in a replicated study in ponderoso pine in the Manitou Experimental Forest in the Colorado Front Range. Good seed crops were produced only every 4 to 6 years, With almost no viable seed produced in intervening years. With seed Predation, only 14% of total seedfall was available for germination. Shelterwood overstories containing between 6 and 14 m(2) ha(-1) stem basal area over scarified seedbeds provided optimal conditions for natural seedling establishment. Survival and growth of planted seedlings was much better than that of natural seedlings. However, poor survival and slow initial growth may require many years to establish a fully stocked forest of natural seedlings.

The Madrean Archipelago, or Sky Island, region of the southwestern United States and northern Mexico is recognized for its great biological diversity and natural beauty. This conference brought together scientists, managers, and other interested parties to share their knowledge about the region and to identify needs and possible solutions for existing and emerging problems. It provided a forum to update the state-of-knowledge acquired since the first conference in 1994. The proceedings contains over 100 articles and additional abstracts from the plenary sessions and from concurrent sessions covering biogeography, ecosystem monitoring, science-based management, cultural resources/history, invasive species, hydrology and biodiversity, conservation planning, ecology, fire, conservation practice, and global climatic change. Abstracts in Spanish are included. The summary of an open forum at the end of the conference provides additional thoughts about current and future needs for the Madrean Archipelago.

In this text we describe an initial attempt to incorporate discrete porous element fuel beds into the coupled atmosphere–wildfire behavior model HIGRAD/FIRETEC. First we develop conceptual models for use in translating measured tree data (in this case a ponderosa pine forest) into discrete fuel elements. Then data collected at experimental sites near Flagstaff, Arizona are used to create a discontinuous canopy fuel representation in HIGRAD/FIRETEC. Four simulations are presented with different canopy and understory configurations as described in the text. The results are discussed in terms of the same two discrete locations within the canopy for each simulation. The canopy structure had significant effects on the balance between radiative and convective heating in driving the fire and indeed sometimes determined whether a specific tree burned or not. In our simulations the ground fuel density was the determining factor in the overall spread rate of the fire, even when the overstory was involved in the fire. This behavior is well known in the fire meteorology community. In the future, simulations of this type could help land managers to better understand the role of canopy and understory structure in determining fire behavior, and thus help them decide between the different thinning and fuel treatment strategies available to them.