Project Title: Calibrating Airborne IR Imagery with Ground-Based Measurements for Describing Fire Behavior and Predicting Fir

The overall objective of this project is to provide wildfire incident teams, resource managers, and scientists with fire behavior and effects maps of wildland fires derived from airborne infrared (IR) overflights. The following specific research and science delivery objectives are proposed.

1. Develop software that produces fire behavior maps from airborne IR imagery. We propose to calibrate IR imagery from the Rochester Institute of Technology's (RIT) Wildland Airborne Sensor Program (WASP) camera with fire behavior data from intensively monitored experimental burns from a wide range of Eastern US fuels (ranging from surface to crown fuels). Continuous airborne IR imagery of developing fires will be obtained. Except for rate of spread, it is currently not possible to derive fire behavior information (e.g., intensity, fuel consumption, residence time, burn/no burn) from airborne IR data because of a lack of calibration equations. Calibration algorithms will be designed to be applicable to any future IR overflight covering the same range of general fire behavior (surface fires in hardwood fuels to crown fires in sand pine scrub).

2. Develop FOFEM-IR, a package of fire effects models for generating fire effects maps from IR overflights. The First Order Fire Effects Model (FOFEM), the primary USFS fire effects model, will serve as the basis for the package. Development will be done in collaboration with ELIZABETH REINHARDT (Missoula Fire Sciences Laboratory) so that FOFEM-IR remains under FOFEM's delivery and support agreement with USFS Fire and Aviation Management. Outputs will include FOFEM outputs (see Background) and new outputs (see Materials and Methods) as justified after consultation with scientists and managers. Inputs to FOFEM-IR will include those for FOFEM, IR overflight data or IR-derived fire behavior maps (Objective 1), and a look-up table of soil and vegetation thermal environments. The look-up table will be developed with the National Institute of Standards and Technology's (NIST) Fire Dynamics Simulator (FDS) for a range of environmental and fire behavior characteristics. FDS simulations will be validated with measurements from the experimental burns.

Once we have calibrated airborne IR imagery to provide fire behavior information and are able to predict key fire effects from those data, we will demonstrate the value of the system by delivering fire behavior and effects maps to wildfire incident teams and our experimental burn cooperators:

3. Science delivery:

a. Rapid delivery of fire behavior and effects maps of burnout operations on wildfires. We will coordinate with REX MANN (Area Commander, Incident Command System) to overfly one or more (nighttime) burnout operations on wildfires. Fire behavior and effects information are needed to assess ecological effects of this fire suppression tool. Rapid map delivery will be accomplished through a system under development by RIT with NASA funding.

b. Deliver airborne IR-derived fire behavior and effects maps to cooperators on experimental burns. Experimental burns associated with IR calibration will be designed to allow us to answer fire behavior and effects questions of interest to our cooperators (see Materials and Methods for details). Experimental burn sites are ordered by expected fire intensity.

i. Daniel Boone National Forest – oak-hickory forest.

ii. Southeastern Piedmont Fire and Fire Surrogates Study site – mixed pine hardwood forest.

iii. Jones Ecological Research Center – southern coastal plain longleaf pine savanna.

iv. The Nature Conservancy (TNC) Albany Pine Bush Preserve Commission and Long Island Chapter – mown, heavy shrub fuels.

v. Myakka River Fire and Fire Surrogates Study site – heavy palmetto fuels in southern flatwoods.

vi. Ocala National Forest – sand pine scrub crown fuels.

c. Deliver fire behavior calibrations and FOFEM-IR for use by airborne IR imaging and GIS groups. Software and updated documentation and training materials will be made available for generating fire behavior and effects maps from future airborne IR flights. Software will be beta tested during the process of producing maps for wildfire incident teams and experimental burn cooperators.

PROJECT JUSTIFICATION

Our project addresses AFP 2005-1, Task 1, by allowing any future airborne IR flight using our algorithms to provide pertinent fire behavior and effects information on wildland fires, whether uncontrolled or prescribed, as they happen. We will demonstrate this capability during the project. Below, we justify our approach to producing fire behavior and effects maps.

Infrared (IR) imagery from airborne platforms (fixed wind aircraft, helicopters, dirigibles) has the potential to provide considerable information while fires are burning. However, the usefulness of IR imagery of fires is seriously limited at present because the IR flux data has not been calibrated to provide fire behavior and effects information beyond rate of spread and relative measures with uncertain meaning (e.g., "fire temperatures", and fire "severity"). Objective 1 directly addresses the calibration problem by allowing IR data to be converted to fire behavior estimates. Our proposal is unique in that we will collect continuous IR imagery of fires as they develop (see Materials and Methods) while past overflights provided discontinuous data, seriously limiting the potential for calibrating the imagery with sparse ground data. Chasing wildfires to obtain calibration information is a highly inefficient use of resources, calibration data should be obtained from intensively sampled experimental burns as we propose. In order that the fire intensity (kW/m, proportional to flame length) range of experiments overlaps that of both wildfires and prescribed burns as much as possible, both surface fires (in hardwood litter; grass; mown scrub oak; and heavy palmetto fuels) and crown fires (sand pine scrub) will be overflown and instrumented.

Although we will use a fixed-wing aircraft in this project (flying a circular pattern with an oblique camera angle), we have discussed using an airship (dirigible) for continuous wildland fire surveillance. Airships are ideally suited to collecting continuous imagery (a potential boon for wildfire incident management and research applications) however current costs are prohibitive (~$10K/day). We have been discussing ways to bring the costs down with STEVE HEWITT (Naval Air Assistance Command, 301-342-8293). A demonstration of airship surveillance of an experimental burn or wildfire would be valuable. We are willing to coordinate a demonstration during this project and would like to discuss (with JFSP and FAM) how we could make that happen. Currently, an airship is sporadically available from Patuxent, MD, within reach of certain of our Eastern US experimental burn sites.

Much recent discussion in the USFS has centered on how to support "one-stop-shopping" for fire behavior and effects models for use by managers. To further that goal, the proposed FOFEM-IR will remain under FOFEM's delivery and support agreement with USFS Fire and Aviation Management (FAM). The FAM agreement does not come with funds, so FOFEM (and FOFEM-IR) must be developed through funds from other sources, such as JFSP. We will modify FOFEM by filling gaps where both a need has been identified through consultation with scientists and managers and a well validated approach exists. Elizabeth Reinhardt, FOFEM developer, will have final say on modifications. Coordination with Reinhardt in FOFEM/FOFEM-IR development will be accomplished by annual meetings and involvement by a software developer under Reinhardt's supervision. New input data sources will include the FDS-generated look-up table of thermal environments, airborne IR data or IR-derived fire behavior maps, and new species information where needed.

Future updates of FOFEM/FOFEM-IR will incorporate more mechanistic fire effects models. A good example of the mechanistic approach is the widely applicable FOFEM soil heating module. Mechanistic models require comprehensive descriptions of soil and vegetation thermal environments. To provide these inputs for new components of FOFEM/FOFEM-IR, and for future updates, we will validate and use a coupled fire-atmosphere model (FDS) to develop a look-up table of soil and vegetation thermal environments (radiation flux and the gas flow velocities and temperatures required to calculate convection heat transfer coefficients). These soil and vegetation thermal environment descriptions are already produced or a logical extension of current FDS outputs. For our purpose of describing soil and vegetation thermal environments, coupled fire-atmosphere models offer considerable benefits in their ability to describe radiant heat fluxes from the fire and air flow in and around both the fire and objects in the fire (e.g., trees). Major limitations to their use include their computational intensiveness (which currently prevents real-time simulations) and a lack of validation data. By using the look-up table approach based on a large number of simulations and simplifying the computations where possible (see Materials and Methods), we can produce needed fire effects output to be used at any future time, thereby circumventing computational limits.

The burns proposed in this project would represent an order-of-magnitude increase in the database of instrumented wildland fires available for validating fire growth predictions from coupled fire-atmosphere models. Ongoing and future projects validating FDS fire growth predictions will also help us towards the goal of predicting fire ignition/extinction (burn/no burn) behavior identified by the Core Fire Science Caucus as a major research need. Ignition and extinction predictions are made routinely during FDS runs.

We will demonstrate our fire behavior and effects mapping capability by delivering airborne IR-derived fire behavior and effects information to wildland fire incident management teams and prescribed/experimental burners in the Eastern US. As requested in AFP 2005-1, Task 1, we will work with our cooperators to design experimental burns so that the burns both meet the requirements for airborne IR data calibration and answer fuel-management related questions our cooperators have identified as important (see Materials and Methods for details). Producing maps will allow us to beta test our software.

BACKGROUND

AIRBORNE IR IMAGING

Infrared imaging has been used in research and operational situations to locate fire perimeters, hot spots and burned areas (e.g., Wilson et al 1971, Matson et al 1981, Hufford et al 1999, Greenfield et al 2003), but has seldom been used in a quantitative fashion to monitor the fine-scale progress of fires, to estimate fire behaviors such as intensity or fuel consumption, or as a tool for fire effects monitoring. Fire behavior (e.g., fire intensity, fuel consumption, burn/no burn) has been difficult to relate to airborne IR imagery because of the infrequency of passes made over fires with the aircraft carrying the cameras. In this project, we propose to collect the needed fire behavior data on fires monitored by nearly-continuous IR imagery. Data within the fire must also be provided to check calibration and atmospheric correction methods for the IR data itself. Such within-fire data has seldom been obtained so absolute calibration of airborne IR instruments has been very difficult. We have developed rugged, in-fire IR flux monitoring capability with radio data links to provide real time ground truth of IR emissions for these calibration needs. In studying wildland fires with airborne IR imagery, it is also desirable to have a sensor with very large dynamic range, so that an unsaturated image of a hot fire may be produced while still having enough sensitivity to observe the relatively cool background radiation emitted or reflected from the earth. With some exceptions, IR instruments have had dynamic range of less than 500, so it has been impossible to observe background radiation without overexposing fire pixels. We have developed a suite of instruments to address these observational problems. The WASP airborne camera, developed with NASA funding and largely with high-performance, commercial off the shelf components, has very large dynamic range (2000 or better) in the short – (1-1.8u) mid- (3-5u) and long-wave (8-11u) infrared, making it ideally suited to fire observations (Kremens et al. 2004a and 2004b).

We have developed a suite of fire intensity and fire perimeter analysis algorithms which will be used to monitor the time history of fire intensity and fire growth (Kremens et al 2004a and 2004b). In experiments conducted in April 2004, we flew the WASP instrument over a 60 acre prescribed fire in Ohio with a sampling interval of 5 minutes, which is adequate to observe slower time-scale fire phenomena such as ground cooling and fire front movement and to obtain a discontinuous "sample" of the fire front over a landscape. Figure 1 shows a fire perimeter map including interior unburned areas from the Ohio fire. In this study, we propose to increase the time resolution to as little as 2 seconds by mounting the camera obliquely in a fixed-wing aircraft and flying a circular pattern. By performing small scale prescribed fire measurements on experimental burns in a variety of terrain and forest types while simultaneously collecting nearly continuous airborne IR imagery, we will ensure that our ground measurements can be used to calibrate and correct the IR imagery and estimate actual fire behaviors from those IR data. Also, we will be able to describe surface heating and cooling rates and determine an optimal mode of operation for producing burn/no burn and fire intensity maps for larger fires using fixed wing aircraft making passes with vertically mounted cameras. As well, by calibrating the continuous IR imagery, we will provide a time-series of fire growth and behavior exceedingly well suited for validating coupled fire-atmosphere models (see below).

Figure 1. Fire perimeter and interior unburned areas after a prescribed fire in an Appalachian mixed-oak forest.

FIRE DYNAMICS SIMULATOR

The Fire Dynamic Simulator (FDS) is a physics-based computer simulation program that models fire behavior and smoke transport. Its development began 25 years ago at the National Institute of Standards and Technology (NIST) with a focus on structural fires (Hostikka and McGrattan 2001, McGrattan 2004). More recently it has been extended to simulate fire behavior in wildland fuels and in the intermix of vegetative and structural fuels occurring at the Wildland–Urban Interface (WUI). The model is one of a class called coupled fire-atmosphere models because it describes the interplay of both combustion and air flow processes (Figure 2). The approach models all modes of heat transfer (radiation, convection, conduction), the gas phase combustion, and the thermal decomposition and pyrolysis of the solid fuel (McGrattan 2004, Hostikka and McGrattan 2001, Baum and Mell 1998, Floyd and Lattimer 2004). Fuels are described by what are called fuel elements. The thermophysical characteristics of the fuel elements (density, surface-to-volume ratio, specific heat, moisture content, etc.) can be measured in the field. These fuel characteristics, along with the net heat flux on the fuel elements, determine the temperature of the fuel elements (which in turn influences the net heat flux). Fire spread experiments in grassland fuels in Australia are currently being used to validate FDS fire behavior predictions and other spread simulations have been done on pine needle litter, shrub fuels, and canopy fuels. Importantly for understanding radiant and convection heat flux to trees in fires (see below), experiments conducted at NIST, in which individual trees and shrubs were burned, are being used to validate FDS predictions of flame spread, heat release rates, and heat fluxes in and around a tree's stem (Evans et al. 2004). More information can be found at: http://www2.bfrl.nist.gov/userpages/wmell/public.html.

Figure 2. FDS simulation of a surface fire spreading from left to right underneath a tree. The thermally thin components of the tree canopy and surface fuels are modeled by fuel elements which are seen in the figure as shaded dots. The fuel elements are shaded according to their temperature, lighter being hotter. Also shown is a shaded contour of the fire plume temperature (lighter being hotter). The section of the tree crown immersed in the fire plume is clearly hotter. Heat transfer into the tree stem and into the soil is not currently modeled.

FIRE EFFECTS MODELING

Our proposed fire effects modeling package (FOFEM-IR) will be based on, and will serve to advance, the First Order Fire Effects Model (FOFEM), a USFS fire effects model widely used by managers (Reinhardt et al. 1997, www.fire.org). The benefit of using FOFEM is that there has been a lot of experience and investment in training associated with the model in accord with its development under a service level agreement with the Washington Office of Fire and Aviation Management for service and delivery. FOFEM currently generates effects predictions in the following categories: tree mortality, fuel consumption (duff, dead woody, herbaceous, shrub, and crown fuels), smoke production, and soil heating. FOFEM is designed to be run in batch mode (meaning that you can run the model rapidly and repeatedly for mapping applications), a feature which will simplify development of FOFEM-IR. We will add modules where there are gaps in FOFEM and where a well validated approach exists and when need has been justified through consultation (see Materials and Methods). As justified by level of validation, new components will be as mechanistic as possible. A mechanistic model is one that, instead of being a statistical relation between variables, is based on the underlying physical, biophysical, and physiological relationships. The problem with statistical models is that they are not valid outside the range of experimental conditions (e.g., range in fire behavior and tree species and sizes) on which they are based and, thus, are often used inappropriately. In contrast, mechanistic models, given proper parameter estimates and if assumptions are adequately met, can be applied more broadly (Dickinson and Johnson 2001). The widely applicable FOFEM soil heating model is a good example of the value of a mechanistic approach. Mechanistic fire effects models, in existence and under development, require that the heat flux or temperature regime (derivable from heat flux) at the surface of the soil or plant component of interest (e.g., stem, branch, bud) is known. These soil and vegetation thermal environments will be provided by a look-up table produced from validated FDS simulation runs. Thermal environments will be produced for a wide range of fire behaviors (e.g., rate of spread, fire intensity) so that ecological effects can be predicted from fire behavior maps derived from airborne IR imagery.

MATERIALS AND METHODS

EXPERIMENTAL BURNS

No ground work will be done on uncontrolled wildland fire incidents, work on wildfires will be limited to airborne IR surveillance and coordinated with REX MANN, Area Commander, Incident Command System (see below). All experimental work will be conducted on prescribed burns managed by cooperators. Dickinson, Kremens, and Bova (as of Spring 2005) have met Firefighter Type 2 training requirements and will attend annual refresher training. Hiring preferences will be for field technicians with firefighter training, but training will be provided for those who do not have it. All project personnel will move to safety zones before a burn is initiated except for qualified personnel assisting with the burn.

Experimental burns for airborne IR calibration and for validating FDS model output will be conducted in single-layer fuels spanning a range of characteristics and expected fire intensity (hardwood litter, grass-dominated pine savanna, pine barrens shrub, palmetto, and sand pine scrub). It will be important to maximize the range in fire behavior in our experimental burns so that our results have the greatest generality. Our goal is to conduct at least 15 burns over all fuel types. Burns in a given fuel type will be conducted in such a way that we obtain not only the calibration and validation data we need but also data needed to address fire behavior and effects questions important to our cooperators conducting the burns. The types of fuels and management questions we will address are summarized below. Sites are ordered roughly in increasing order of expected fire intensity.

Oak-hickory forest. Contact: REX MANN, Staff Officer for Timber, Wildlife, and Fire, Daniel Boone National Forest, Winchester, KY, Tele: 859-745-3123. On Federal lands in the central US, burning during the growing season is strictly limited by the USFWS because of concerns about harming neonate and adult female endangered Indiana Bats in maternity roosts. Of interest is the link between fire intensity and risk to roosting bats. Risk algorithms will be developed as part of a separate project proposed to JFSP.
Mixed pine-hardwoods (Southeastern Piedmont Fire and Fire Surrogates Study [FFS]). Contact: TOM WALDROP - USFS, Southern Research Station, Clemson, SC, Tele: 864-656-5054. Maps of stem kill on the FFS site will be of particular interest. As well, fire behavior is a key FFS variable that, in practice, has been difficult to quantify adequately. The fire behavior maps we propose to derive from our IR overflights would be valuable to projects such as the FFS.
Upland long-leaf pine savanna. Contact: ROBERT MITCHELL – Jones Ecological Research Center, Newton, GA, Tele: 229-734-4706. Silvicultural prescriptions call for cutting overstory pines so as to create gaps to reduce light and belowground competition for pines. However, gaps may have the unintended consequence of promoting hardwood growth and reducing fire intensities. Of interest is fire behavior and hardwood stem kill across experimental thinning treatments in which overstory pines have been cut either in groups or evenly across a treatment.
Pitch pine scrub oak barrens. Contacts: NEIL GIFFORD - The Nature Conservancy, Albany Pine Bush Preserve Commission, Tele: 518-785-1800 ext. 214. BILL PATTERSON - The Nature Conservancy, Long Island Field Office, Tele: 631-367-3384 ext 126. National Fire Plan supported fuel treatments involve mowing heavy shrub fuels and then summer burning some 6 weeks afterwards. Of interest are fire intensity, duff consumption, soil sterilization, and scrub oak and pitch pine mortality.
Southern flatwoods (Myakka River FFS Study site). Contact: KENNETH OUTCALT - USFS, Southern Research Station, Athens, GA, Tele: 706-559-4309. Fire suppression in flatwoods longleaf pine sites has resulted in heavy accumulations of palmetto and other shrub fuels in many areas. These unnatural accumulations of fuels result in high fire intensities that can wound overstory pines. With repeated burning, fire scars enlarge and ultimately result in stem death. Mortality of longleaf pines is a concern at Myakka River because it is at the southern edge of the range of longleaf pine.
Sand pine scrub. Contact: GEORGE CUSTER, Fire Management Officer, Ocala National Forest, FL, Tele: 352-669-3153, Email: gcuster@fs.fed.us. These intense stand-replacing crown fires will provide the opportunity to increase the range over which we calibrate the airborne IR imagery for fire behavior. We will provide fire behavior and other ecological effects maps as requested.

In-fire instruments will be distributed over the burn units depending on fuel characteristics and the need for the fire to reach quasi-steady state as it spreads past the instruments (Figure 3). An approximately linear fire line at the instruments is preferred. We will work with our cooperators to meet these objectives. The in-situ data loggers have been used to record surface temperatures and decay time of ground cooling after fires and to measure fire weather (temperature, relative humidity, wind speed and direction in the horizontal plane) and time-resolved infrared flux and emissivity (e.g., Kremens et al. 2003b). For our experiments, additional sensor packages will be provided to measure the mid-flame and 20 foot wind speed and direction in the horizontal plane, vertical wind speed, the infrared flux emitted from the soil surface in two bands (0.1- 10 μm and 6-13 μm), and the downwelled visible light (for surface heating effects). For observation of gross fire behavior near critical instrument stations we will deploy several (2-3) custom wireless-controlled in-fire video cameras that we have built for past projects. All of these modifications and additions have been developed under other funding sources and will be applied rapidly to this project. A summary of the in-fire instrumentation is given in Table 1.

Table 1 – Variables to be measured by in-fire instruments by application.

Variable	Use	Instrument
Mid-flame and 20' wind speed	FDS validation/parameterization	Cup anemometer
Mid-flame and 20' wind direction	FDS validation/parameterization	Wind vane
Relative humidity	FDS validation/parameterization	Electronic sensor
Air temperature	FDS validation/parameterization	Electronic sensor
Mid-flame and 20' vertical wind speed	FDS validation/parameterization	Differential pressure transducer
Soil temperature	Soil heating model	Thermocouples (2-4)
Surface infrared flux	Airborne image calibration, soil heating model	Thermopile radiometer (1-2)
Visual fire behavior (e.g., flame lengths, leeward vortices)	FDS validation/parameterization	Fire hardened video cameras

Software: Microsoft Office

Figure 3 – Rapid-deploy in-fire data logger and mid-flame instrument mast for wildland fire research. The logger box and mast are shielded to enable burnover. Some inexpensive instruments are destroyed by flame contact depending on flame length and mast height. Instruments for in-fire deployment are listed in Table 1. Photos: Robert Kremens.

Additional instruments will be required to estimate heat flux to tree stems to provide validation data for FDS stem heating simulations. First, water-cooled radiant and total flux sensors will be inserted flush with the tree stem (Jones et al. 2004a and 2004b). The surface of the flux sensor housing will be instrumented with a thermocouple in order to estimate radiation loss and better constrain convection heat transfer (total – radiant heat flux). Second, temperature traces from thermocouples placed at known depths within a stem and an inverse computational method will be used to estimate net heat flux (total – radiant heat flux) into the stem. Multiple sets of thermocouples will be placed around each stem to quantify the uneven heating. We will work with BRET BUTLER (USFS, Missoula Fire Science Lab, Tele: 406-329-4801) to accomplish these tree stem heating measurements. Butler's cooperation will be valuable because of his experience and equipment from a past JFSP-funded stem heating project (on which Dickinson and Bova collaborated). If funded, data will be shared between this project and Butler's proposed 2005 JFSP FireStem project.

Replicate measurements of fuel consumption and rate of spread are required in association with in-fire instrumentation. Rate of spread will be derived from the IR imagery. Fuel consumption will be determined by destructive and non-descructive sampling. Destructive sampling will be conducted in a representative area somewhat away from instrument packages. Standard fuel sampling methods will be employed (e.g., FIREMON protocols). Fire intensity (kW/m) will be estimated from fuel consumption, rate of spread, and standard assumptions about fuel heats of combustion and fuel fraction involved in flaming combustion. We will use IR imagery (both airborne and in-fire), measurements of flaming combustion residence times (Bova and Dickinson 2004a), and FDS estimates to constrain our field estimates of heat release in the main fire front.

AIRBORNE IR IMAGING

Airborne IR imagery will be collected continuously on the experimental burns by flying a circular pattern with the WASP camera mounted at an oblique angle. We will choose sites for instrumentation and flight altitude so that the airborne IR imagery encompasses both the instrumented area and other areas of interest to our cooperators. We will capture images in three infrared bands (SWIR, MWIR and LWIR) as well as RGB color or color-IR. These data will be analyzed for fire intensity after point calibration of IR flux with 3-10 in-situ data loggers. Calibration of the WASP IR data removes effects of the atmosphere and fire-generated smoke. Corrected IR imagery will be further calibrated to produce fire behavior estimates with in-fire instrument data, ground-based fire behavior estimates, and FDS fire simulation results. By assembling a collection of calibration data over various intensities and types of fires (with varying smoke conditions) calibration equations will be developed for future wildfire events. We will use and modify our suite of suite of fire intensity and fire perimeter analysis algorithms to describe the time history of fire intensity and growth.

FIRE DYNAMICS SIMULATOR

The Fire Dynamics Simulator (FDS) will be used to simulate, for a representative range of environmental (slope and wind) and fire (heat release rate and fire line depth) conditions, the thermal environment (in terms of heat fluxes and gas velocities and temperatures) of soils, tree stems, and canopy elements (buds, branches, foliage) in prescribed fires. The considerable output of the simulations will be organized in a matrix (the look-up table) in which each cell is defined by its environmental and fire conditions. FDS is capable of predicting the time-dependent spread of a fire through a three-dimensional array of surface and suspended vegetative fuels. However, if the experimental firelines have a sufficiently linear shape and constant spread rate it may be possible to realistically represent the fire as a fixed linear burner of infinite length. This two-dimensional approximation would lead to a number of simplifications and computational savings. For example, the time history of the incident radiant flux on a given target (e.g., a portion of the soil or canopy) could be determined from the averaged computed incident radiant heat flux, qR,i(x,y,z). The distance x, parallel to the direction of fire spread, can be related to the time through t = |x – xo| / R, where R is the constant spread rate of the prescribed fire and xo is the original location of the target. The resulting time history of incident radiant flux would be used in models for soil, tree stem, and canopy heating. It is expected that the bulk of canopy heating (at least for surface fires) and much of the tree stem heating will be through convection. Accordingly, gas velocities and temperatures, and their time courses, will be computed for a range of heights above ground to enable convection heat-transfer coefficients to be estimated for stems and canopy elements.

The above two-dimensional approach is most appropriate for simulating the thermal environment of the soil and of smaller sized fuels (buds, branches, foliage) which are assumed to be homogeneously distributed in space. It should not be used, for example, for tree stems of sufficiently large diameter that the flow of heated gases and flames interact to create vortices that preferentially heat their leeward side. In this case the full transient, three-dimensional implementation of FDS would be needed. In order to ensure consistency with the static two-dimensional simulations, a spreading line fire with the same constant spread rate and fire depth will be used in the three-dimensional simulations. Three-dimensional simulations using fuel characteristics measured in the field (bulk densities, packing ratio, size and spatial distribution of tree stems, etc.) will be used to determined under what conditions the simplified two-dimensional modeling approach is appropriate.

One dimensional models of the thermal degradation of wood are currently part of FDS and predict heat up, moisture evaporation, pyrolysis, and char formation during thermal degradation. These models, along with a tissue necrosis model, will be used to produce estimates of vascular cambium kill depth for comparison with the stem kill module in the ecological effects package (FOFEM-IR).

FIRE EFFECTS MODELING

The First Order Fire Effects Model (FOFEM, www.fire.org), on which FOFEM-IR will be based, is widely used by managers. We will collaborate with ELIZABETH REINHARDT, the FOFEM developer (USDA Forest Service, Missoula Fire Sciences Lab, Tele: 406-329-4760), from the beginning of the project to make sure our modifications are most conducive to advancing FOFEM itself and are consistent with the service level agreement between Reinhardt and the Washington Office of USFS Fire and Aviation Management for FOFEM delivery and support. Reinhardt will have final say on FOFEM/FOFEM-IR development. Proposed additions to FOFEM/FOFEM-IR are described below. The airborne IR-derived fire behavior maps and the thermal environment look-up table produced by FDS will be important new sources of input data for the ecological effects package. Programming will be conducted in C in accord with FOFEM and coordinated through the participation of a programmer working under Reinhardt's supervision.

Species characteristics databases will be added as needed (e.g., branch dimensions for the canopy branch necrosis model) and existing FOFEM databases will be augmented as needed with species of importance to our cooperators on this proposal. Additions to FOFEM/FOFEM-IR will be determined by availability of well validated approaches and consultation. Consultation will include feedback on FOFEM received continuously by Reinhardt and a consultation process with scientists and managers to be conducted at the outset of the project. Additions may include a stem heating and vascular cambium necrosis routine for predicting stem kill and wounding (Bova and Dickinson 2004a or, if 2005 JFSP proposal is funded, FireStem [Jones et al. 2004a and 2004b, www.fire.org]), a canopy branch necrosis routine that will predict proportional crown kill (Mercer et al. 1994), a serotinous cone opening routine (Johnson and Gutsell 1994), a soil hydrophobicity routine (Wells et al. 1978), and a soil microbial necrosis routine that will be linked with the FOFEM soil heating model to predict sterilization/root-kill depths. The microbial necrosis routine will be based on an extensive microbial heat-effects literature (see review in Dickinson and Johnson 2001). The stem-kill module will be written to predict not only complete vascular cambium necrosis around a stem but also the partial vascular cambium necrosis that produces fire scars, of interest in the Myakka River FFS site (see below). For this fire wounding application, and for the stem mortality routine in general, it is critical to have the circumferential FDS-derived heat fluxes for tree stems.

WILDFIRE BURNOUT OPERATION

In the 3^rd year of the project, we will work with REX MANN, Area Commander with the Incident Command System (Daniel Boone National Forest, Winchester, Kentucky, Tele: 859-745-3123) to identify one or more large wildfires (probably in the Western US) for overflight. The objective will be to assess the fire behavior and effects associated with burnout operations in which fires are set, often aerially and at night, in order to consume fuels between a control line and the wildfire perimeter. Burnout operations are a relatively economical tactic for fighting the large fires that account for the greatest portion of fire suppression and control expenditures. There is a need for information on these fires so that their ecological effects can be assessed and, potentially, mitigated. If possible, the overflights will be coordinated so that continuous imagery of the fire can be obtained (by using either an airship or a fixed-wing craft flying a circular pattern with an obliquely mounted camera). Rapid map delivery will be accomplished through a system under development by RIT with NASA funding. If the overflight(s) occur late in Year 3, we expect to be able to make same-day delivery of the fire behavior and effects maps.

SCIENCE DELIVERY AND APPLICATION

Airborne IR fire behavior calibrations and the thermal environment look-up table will be incorporated in the FOFEM-IR software package for implementation by airborne IR and GIS groups that produce maps for research, management, or wildfire suppression and control. The spatial resolution of the results will be dependent on how frequently the plane passes over the fire, the plane's altitude, and the resolution of the camera. The FOFEM-IR software and updated documentation and training materials will be produced in collaboration with ELIZABETH REINHARDT and with assistance from a programmer working under her supervision. All materials will be distributed through www.fire.org (maintained by the USFS Missoula Fire Science Lab). FOFEM-IR will be included with FOFEM under the support and delivery agreement with Fire and Aviation Management.

We will demonstrate our fire behavior and effects outputs (and beta test our software in the process) by delivering requested maps to our cooperators on prescribed burns and wildfire burnout operations. Maps to be produced for our cooperators include: burn/no burn, fire intensity, flame length, rate of spread, fuel consumption, duff consumption, soil sterilization depth, potential stem mortality (based on bark thickness and species), and canopy tree stem wounding (based on tree size, wind, and bark thickness).

An important dimension of our wildfire burnout operation overflight(s) will be rapid delivery of information. At present, RIT is capable of delivering same-day wildfire perimeter information. At the end of the proposed project, rapid delivery of fire behavior and effects maps will be possible.

COOPERATORS AND LINKAGES WITH OTHER PROJECTS

The following people are Cooperators on this project. Their contributions and contact information are noted in the text where appropriate. We will coordinate with REX MANN¹ (USFS, Incident Command System and Daniel Boone National Forest) on both the wildfire burnout operation overflight(s) and on experimental burns. We will coordinate with NEIL GIFFORD¹ (TNC), BILL PATTERSON (TNC), GEORGE CUSTER (USFS, Ocala National Forest), KENNETH OUTCALT¹ (USFS, Southern Research Station), TOM WALDROP (USFS, Southern Research Station), ROBERT MITCHELL¹ (Jones Ecological Research Center) on experimental burns that they will be conducting during the first 1.5 years of the project. We will collaborate with ELIZABETH REINHARDT¹ (USDA Forest Service, Missoula Fire Science Lab) on FOFEM/FOFEM-IR development. We will collaborate with BRET BUTLER (USDA Forest Service, Missoula Fire Science Lab) in-fire instrumentation needs and methods.

¹Support letter attached.

This project will link with other ongoing and proposed projects.

Wildland Airborne Sensor Program (WASP) I and II (2003/2004 NASA project, NAG13-02051 and NAG13-03026). NASA projects developed the IR imaging systems to be used in this project. Integrated Sensing Systems Initiative (ISSI) (rapid deployment in-fire instrumentation) (2004 NASA project). Initial development of the instrumentation to be used in monitoring experimental burns has been developed in previous NASA (Forest Fires Imaging Experimental Systems NAG5-10051) and JFSP (see next) projects. This equipment measures IR surface Flux in multiple bands, ground temperature monitored by thermocouples, and in-fire weather (wind speed, direction, RH, dew point and temperature). The ISSI project will expand the capability of these devices to include remote radio reporting and other optical and smoke/particulate monitoring.
Will provide fire behavior and effects maps for two JFSP funded Fire and Fire Surrogates study sites and for two TNC fuel treatment projects supported by the National Fire Plan (see Materials and Methods for details).
Rapid delivery of fire maps (WASP II 2004 NASA, NAG13-03026). This project is developing the hardware and software capability for rapid (in-flight) processing and delivery of fire maps derived from airborne IR data. Fire behavior calibration algorithms developed in this project will enable delivery to fire incident teams not only of IR imagery, but also estimated fire behavior.
Fire database development project (JFSP 2003 Rapid Response Project, ‘Demonstration and Integration of Systems for Fire Remote Sensing, Ground-Based Fire Measurement, and Fire Modeling’ C. Hardy and P. Riggan). As part of this project, Lloyd Queen and collaborators at the University of Montana developed a time-resolved, multi-layer GIS database for ground based, oblique view, and overhead images and point data. Future plans would involve using this database structure to enable easy access to our data by our team and other researchers.
Past JFSP grant (and 2005 JFSP proposal) to Bret Butler, Missoula Fire Sciences Lab, for development of FireStem (www.fire.org). Dickinson and Bova cooperated with Butler and colleagues on past FireStem work and Dickinson contributes to Butler's current JFSP proposal through his tree vascular cambium thermal tolerance work. We will work with Butler to maximize field validation of FireStem.
Proposal to JFSP (2005, Dickinson and others, "Injury and Mortality Risks from Wildland Fire Smoke and Heat Exposures for Endangered Indiana Bats [Myotis sodalis] in Maternity Roosts"). This project will develop modeling tools to determine bat mortality and injury from smoke and heat exposures above wildland fires. If funded, this project will enable us to convert our fire intensity maps to maps of threshold heights for bat death and injury from smoke and heat.
Proposal to JFSP (2005) on fire behavior in the wildland-urban interface (National Institute of Standards and Technology and USFS) the data collected is this airborne IR calibration project will be a crucial complement to data collected in the proposed WUI project.

PROJECT DURATION, ACTIVITIES, REPSPONSIBILITIES, AND TIMELINE

The expected duration of this project is three years from the initiation of funding (Table 2). The timelines for individual project components that will ultimately allow us to deliver fire behavior and effects maps are given in Table 2. It is important that experimental burns and concomitant overflights be conducted during the first half of the project. Experience at RIT and the Northeastern Research Station (NERS) on wildland fire overflights and in-fire sampling and instrumentation will allow us to begin this work quickly. Also, the differences in seasonality of prescribed burning among our Cooperators' sites will facilitate the project.

Refereed journal publications will serve as the foundation of the airborne IR fire behavior calibrations and FOFEM-IR and will be produced both on the individual project components of the project and on the synthesis of the different components of the project required to produce the fire behavior and ecological effects maps (see Deliverables, below).

Table 2 - Activities, group with primary responsibility, and timeline. Acronyms not defined above: York University (YU).

Task	Primary Responsiblity	Year 1	Year 2	Year 3
Experimental burn instrumentation and airborne IR overflights	RIT/NERS
Calibrate IR overflight data with experimental burn data to produce fire behavior maps	RIT/NERS
FDS configuration for thermal environment simulation runs	NIST/YU
FDS validation/calibration against experimental data	NIST/YU
Use FDS to generate look-up table for soil and vegetation thermal environments (for use in FOFEM-IR)	NIST/YU
Develop ecological effects modeling package (FOFEM-IR) based on FOFEM, updated as needed and justified	NERS
Link FOFEM-IR package with FDS look-up table and airborne IR-derived fire behavior maps	NERS
Generate fire behavior and effects maps for cooperators (beta test software)	RIT/NERS
Wildfire burnout operation overflight, rapid delivery of fire behavior and effects maps (beta test software)	RIT/NERS
Production of airborne IR fire behavior calibrations and FOFEM-IR software packages	NERS/RIT

DELIVERABLES

Deliverables are summarized in Table 3. Annual reports will be delivered by February 15 each year. The final report will be delivered by the project termination date (end-Year 3). Dickinson will be responsible for these reports. Refereed publications will also be produced from each component in Table 2, but are not shown. Software and documentation will be made available on the web through the Missoula Fire Sciences Laboratory (www.fire.org). Apart from annual and final reports, the following deliverables are proposed.

Table 3 – Deliverables.

Deliverable	Description	Delivery Date(s)
Airborne IR fire behavior calibration software	Statistical and physically-based equations will be produced that will enable anyone with GIS capabilities to estimate fire behaviors from IR overflights.	End-Year 2
Look-up table of soil and vegetation thermal environments	FDS will be used to produce a database (look-up table) of soil and vegetation thermal environments for use with FOFEM-IR.	Mid-Year 3
Ecological effects modeling software (FOFEM-IR)	A FOFEM-based ecological effects modeling package designed to be linked with airborne IR fire behavior output and the FDS look-up table. Collaboration with Reinhardt at Missoula Fire Sciences Lab.	Mid-Year 3
Fire behavior and ecological effects maps for prescribed burn cooperators	Production of fire behavior and effects maps requested by cooperators	End-Year 3
Wildfire burnout operation fire behavior and effects maps	Assessment of the ecological effects of burnout operations on wildfires. If overflight(s) occur after other work is completed, information delivery will be same-day.	End-Year 3

BUDGET

The budget can be roughly broken down into support for producing fire behavior maps from airborne IR imagery (experimental burns, airborne IR overflights, and associated data analysis; ~65% of costs) and support for producing FOFEM-IR (developing FOFEM-IR and FDS modeling; ~35% of costs). Not included in the estimate for FOFEM-IR development is the portion of the instrumentation that will be tailored to FDS model validation needs. As well, FDS will help us constrain our fire intensity estimates, a substantial benefit in the effort to produce fire behavior calibrations from IR imagery. As such there is synergy between project components and the costs can not be completely separated. Because of the need for experimental burn validation data for both parts of the project, conducting the fire behavior and effects parts together will cost less than conducting them separately.

Mell's position at the National Institute of Standards and Technology (NIST) is not permanent and is supported by grant funds. Most of the FDS simulations will be conducted by a Jenkins graduate student at York University. However, Mell's contributions to the FDS work on this project will be indispensable because of his long involvement in FDS development at NIST and the new FDS applications we propose. Accordingly, inclusion of his salary and indirect costs are necessary to accomplish the project objectives.

Total value of in-kind and financial contributions is well over $438K. Only figures labeled below as "contributed" are included in this total. No attempt was made to determine contributed costs for already-purchased equipment and software, these costs will not be trivial (see below). A portion of RIT salary costs are not included in order to reduce the project's cost and no contributed salary costs are included for York University. Other contributions follow: USFS, Northeastern Research Station - Scientist (0.5 FTE for three years)(contributed ~$120K); Project Manager (0.2 FTE for three years)(contributed ~$25K); NERS field equipment (thermocouples, loggers, total cost ~$15K). Rochester Institute of Technology - 3-year tuition costs for MS student (contributed ~$71K); 22% of indirect costs (current rate is 42%)(contributed ~$91K); WASP instrument and non-renewable equipment construction (total cost $600K); 20 field sensor boxes ($10K + NRE ~ $50K total cost); ENVI, Imagine and IDL software suites (~$20K total cost). National Institute of Standards and Technology - 59% of indirect costs (current rate is 79%)(contributed ~$74K). York University - 42% of indirect costs (current rate is 62%)(contributed ~$32K). Miscellaneous - Costs for calibration burns are borne by cooperators. If we assume $5K per collaborator and 5 collaborating groups (contributed ~$25K).

Table 4 – Project budget. Amounts are in thousands of dollars ($1000 = $1K).

	Year 1	Year 2	Year 3	Total
USFS Northeastern Research Station, Project 4153
Salary - Term Physicist, 0.8 FTE	$48.0K	$49.6K	$50.4K	$148.0K
Salary - GS5, Field Technician, 1 FTE	30.0	31.0		61.0
Salary - GS5, Field Technician, 1 FTE	27.0	28.0		55.0
Travel - PI meeting	1.0	1.0	1.0	3.0
Travel - field and meetings	8.0	8.0	2.0	18.0
Equipment - thermocouple manufacture, sensors	4.0			4.0
Supplies - consumable materials, wires, shielding	1.0	1.0		2.0
Cooperative and Other Agreements
Rochester Institute of Technology
Salary - Sr. Scientist, 0.33 FTE	38.9	40.3	42.2	121.4
Salary - Project Manager, 0.04 FTE	6.9	7.0	7.2	21.1
Salary - Engineer, 0.33 FTE	19.1	20.5	21.1	60.7
Salary - MS Student	29.0	29.0	29.0	87.0
Travel - field and meetings	8.0	8.0	3.0	19.0
Equipment - next generation ground sensors	10.0	2.0		12.0
Supplies - consumable materials, wires, shielding	10.0	10.0		20.0
Instrument maintenance	10.0	10.3	2.5	22.8
Flight costs - $500/hr, minimum 40, 40, 10 hours	22.0	24.0	5.0	51.0
Indirect costs (20%)	30.8	30.2	22.0	83.0
National Institute of Standards and Technology
Salary - Sr. Scientist, 0.33 FTE (w/fringe benefits)	40.0	41.0	42.0	123.0
Travel - field and meetings	1.0	1.0	1.0	3.0
Indirect costs (20%)	8.2	8.4	8.6	25.2
York University
Salary - PhD Student	18.5	18.5	18.5	55.5
Travel - field and meetings	3.0	3.0	2.0	8.0
Equipment - computers (2)	10.0			10.0
Supplies and publication costs	1.0	1.0	1.0	3.0
Indirect costs (20%)	6.5	4.5	4.3	15.3
Missoula Fire Sciences Lab, Fire Effects Project
FOFEM-IR software development and delivery support	5.0	5.0	5.0	15.0
Travel - FOFEM-IR coordination meetings	1.0	1.0	1.0	3.0
Northeastern Research Station Indirect Costs
On passthrough funding (10%)	27.3	25.9	20.9	74.1
On direct Project 4153 costs (15%)	17.9	17.8	8.0	43.7
Total Per Year and for Project	443.1	426.9	297.8	1167.9

LITERATURE CITED

Baum, H. R. and Mell, W. E. 1998. A radiative transport model for large-eddy fire simulations. Combustion Theory Modelling 2:405-422.

Bova, A. S., M. B. Dickinson. 2004a. Beyond “fire temperatures”: using thermocouple probes to estimate surface fire intensity and fuel consumption. Canadian Journal of Forest Research. In review.

Bova, A. S., and M. B. Dickinson. 2004b. Linking surface fire behavior, stem heating and tissue necrosis. Canadian Journal of Forest Research. Accepted.

Campbell, G. S., Jungbauer, J. D., Jr., Bidlake, W. R., and Hungerford, R. D. 1995. Soil temperature and water content beneath a surface fire. Soil Science 159(6):363-374.

Dickinson, M. B., and E. A. Johnson. 2001. Fire effects on trees. In E. A. Johnson and K. Miyanishi (eds) Forest Fires: Behavior and Ecological Effects. Academic Press.

Dickinson, M. B., and E. A. Johnson. 2004. Temperature-dependent rate models of vascular cambium cell mortality. Canadian Journal of Forest Research 34:546-559.

Dickinson, M. B., Jolliff, J., Bova, A. S. 2004. Vascular cambium necrosis in forest fires: using hyperbolic temperature regimes to estimate parameters of a tissue-response model. Australian Journal of Botany. In press.

Evans, D. D., Rehm, R. G. and Baker, E. S. 2004. Physics-Based Modeling for WUI Fire Spread - Simplified Model Algorithm for Ignition of Structures by Burning Vegetation. NISTIR 7179, 2004

Floyd, J. and Lattimer, B. 2004. Validation of FDS V4 Boundary Heat Flux Predictions for a Corner Fire. Interflam 2004: 10^th International Fire Science and Engineering Conference, Edinburgh, Scotland, July 5-7, 2004, pp. 1281-1292.

Greenfield, P., Smith, W., Chamberlain, D., PHOENIX – The new Forest Service airborne infrared fire detection and mapping system, published in the proceedings of the 2^nd Fire Ecology Congress (American Meteorological Society) Orlando Fl, November 2003

Hostikka, S. and McGrattan, K.B.. 2001. Large Eddy Simulation of Wood Combustion. International Interflam Conference, 9th Proceedings, September 17-19, 2001 pp. 755—762

Hufford, G. Kelley, H., Moore, R., Cotterman, J., 1999, Detection and growth of an Alaskan forest fire using GOES-9 3.9 mm imagery, Intl. J.of Wildland Fire, 9:129-136

Johnson, E. A., and S. L. Gutsell. 1993. Heat budget and fire behavior associated with the opening of serotinous cones in two Pinus species. Journal of Vegetation Science 4:745-750.

Jones, J. L., Webb, B. W., Butler, B. W., Dickinson, M. B., Jimenez, D., Reardon, J., Bova, A. S. 2004a. Prediction of Thermally-Induced Stem Mortality in Fires. Canadian Journal of Forest Research. In review.

Jones, J.L., Webb, B.W., Jimenez, D., Reardon, J., and Butler, B. 2004b. Development of an advanced one-dimensional stem heating model for application in surface fires. Can. J. For. Res. 34:20-30.

Kremens, R., Dickinson, M. B., Bova, A. S, Faulring, J. 2004a. Accurately mapping unburned areas using time-sequenced airborne imaging. Assoociation for Fire Ecology Mixed Severity Fire Regime Conference, Spokane WA

Kremens, R., Bova, A., Dickenson, M., Faulring, J., Hardy, C., 2004b, Determining the granularity and randomness of burned areas vs. unburned areas from prescribed fires, to be presented at the International Association of Landscape Ecology, Syracuse, NY

Kremens, R., Faulring, J., Gallagher, A., Seema, A., and Vodacek, A. 2003a. Autonomous field- deployable wildland fire sensors. Intl. J. of Wildland Fire 12:237-244.

Kremens, R., Faulring, J., Hardy, C., 2003b, Measurement of the time-temperature and emissivity history of the burn scar for remote sensing applications, published in the proceedings of the 2^nd Fire Ecology Congress (American Meteorological Society) Orlando Fl, November 2003.

Matson, M. and Dozier, J. 1981. Identification of subresolution high temperature sources using a thermal IR sensor, Photo. Engr. and Rem. Sens., 47, 1311-1318.

McGrattan, K. B. 2004. Fire Dynamics Simulator (Version 4), Technical Reference Guide, NISTIR Special Publication, 1018.

Mercer G. N., Gill A. M., & Weber R. O. 1994. A time-dependent model of fire impact on seed survival in woody fruits. Aust J Bot 42, 71-81.

Wells, C. G., Campbell, R. E., DeBano, L. F., Lewis, C. E., Fredriksen, R. L., Franklin, E. C., Froelich, R. C., and Dunn., P. H. 1979. Effects of fire on soil, a state-of-knowledge review. USDA Forest Service, Washington Office, General Technical Report WO-7.

Wilson, R., Hirsch, S., Madden, R. 1971. Airborne infrared forest fire detection system: Final report, USDA Forest Service Research Paper INT-93.

QUALIFICATION OF INVESTIGATORS

Following are CVs from Dickinson, Kremens, and Mell. Summarized biographical sketches are also included for Jenkins, McKeown, and Bova.

MATTHEW B. DICKINSON

Education

1998 Ph.D., Biology (Ecology and Evolution), Florida State University

1991 M.S., Biology (Ecology and Evolution), Florida State University

1988 B.S., Marine Biology, Texas A&M University, Galveston (Magna Cum Laude)

Professional Experience (since 1995)

2001-present Research Ecologist, USDA Forest Service

1997-2001 Postdoctoral Associate, University of Calgary

1994-1995 Research Associate, Smithsonian Environmental Research Center (in Mexico)

1995-1996 Research Assistant, Tall Timbers Research Station

Research Interests

Tree mortality in wildland fires, mechanistic modeling of ecological processes, forest fire extinction, forest fire spread and effects in a landscape context, ecosystem process modeling.

Selected Publications and Presentations

Bova, A. S, Kremens, R., and Dickinson, M. B. 2004. Calibrating aerial IR imagery with ground-level observations of a prescribed burn for predicting ecological effects. Ecological Society of America, 89th Annual Meeting, Portland, OR. CD-ROM p.57.

Bova, A. S., and Dickinson, M. B. 2004. Linking surface fire behavior, stem heating and tissue necrosis. Canadian Journal of Forest Research, accepted.

Bova, A. S., and Dickinson, M. B. 2004. Estimating fire intensity and fuel consumption from thermocouple probe response to surface fires. Canadian Journal of Forest Research, in review.

Dickinson, M. B., and E. A. Johnson. 2004. Temperature-dependent rate models of vascular cambium cell mortality. Canadian Journal of Forest Research 34:546-559.

Dickinson, M. B., Jolliff, J., and Bova, A. S. 2004. Vascular cambium necrosis in forest fires: using hyperbolic temperature regimes to estimate parameters of a tissue-response model. Australian Journal of Botany, in press.

Jones, J. L., Webb, B. W., Butler, B. W., Dickinson, M. B., Jimenez, D., Reardon, J., and Bova, A. S. 2004. Prediction and Measurement of Thermally-Induced Mortality in Tree Stems. Canadian Journal of Forest Research, in review.

Wildman, R. A., Hickey, L. J., Dickinson, M. B., Berner, R. A., Robinson, J. M., Dietrich, M., Essenhigh, R. H., Wildman, C. B. 2004. Burning of forest materials under late Paleozoic high atmospheric oxygen levels. Geology 32(5):457-460.

Yaussy, D. A., M. B. Dickinson, and A. S. Bova. 2004. Prescribed surface-fire tree mortality in southern Ohio: equations based on thermocouple probe temperatures. Proceedings of the 14^th Central Hardwoods Forest Conference, March 16-19, Wooster, Ohio. USDA Forest Service, General Technical Report NE-316. P. 67-75.

Dickinson, M. B. 2002. Heat transfer and vascular cambium necrosis in the boles of trees during surface fires. In : X. Viegas (ed.) Forest Fire Research & Wildland Fire Safety, Millpress, Rotterdam.

Dickinson, M. B., and E. A. Johnson. 2001. Fire effects on trees. In E. A. Johnson and K. Miyanishi (eds) Forest Fires: Behavior and Ecological Effects. Academic Press.

Gutsell, S. L., E. A. Johnson, K. Miyanishi, J. E. Keeley, M. B. Dickinson, and S. R. Bridge. 2001. Varied ecosystems need different fire protection. Letter, Nature 409:977.

Miyanishi, K., E. A. Johnson, S. L. Gutsell, M. B. Dickinson, R. D. Revel. 2000. Lightning fires. Research Links 8(3):18-19.

Dickinson, M. B., and E. A. Johnson. Predicted spread and extinction of surface fires in aspen and conifer fuels in the Canadian mixedwood boreal forest. Ecological Society of America Annual Meeting, Snow Bird, Utah. August 2000.

Dickinson, M. B. Fire effects on trees. National Center for Ecological Analysis and Synthesis, Santa Barbara, California. April 1999.

Professional Societies: Ecological Society of America, American Association for the Advancement of Science, International Association for Fire Ecology.

Competitive Grants

Agenda 2020 Program, Northeastern Research Station, $120,000, " Mapping Landscape Forest Canopy Structure with High Resolution Satellite Imagery", with Dr. Conghe Song, 2005-2007.

National Science and Engineering Research Council of Canada, National Centers of Excellence, Sustainable Forest Management Network, $45,000, "Fire Ignition and Extinction in the Mixed-Wood Boreal Forest", with Prof. E. A. Johnson, 1999-2001.

US Man and the Biosphere Program, $20,000, with D. F. Whigham, N. V. L. Brokaw, and L. Poot-Chan, for work on mahogany regeneration in Mexico and Belize, 1995-1997.

Smithsonian Institution, $36,000, for work on tree regeneration and bird response to logging in Quintana Roo, Mexico, 1992-1995.

ROBERT KREMENS

Professional Preparation

2000 M.S. Environmental Studies, University of Rochester, Rochester, NY

1981 Ph.D. Physics, New York University, New York, NY

1977 M.S. Physics, New York University, New York, NY

1975 B.S., Physics, The Cooper Union, New York, NY

Appointments

2000 - present Senior Research Scientist, Rochester Institute of Technology

1998 - 2000 Senior Principal Consultant, Questra Consulting

1986 - 1998 Scientist, Laboratory for Laser Energetics, University of Rochester

1985 - 1986 Design Engineer, LeCroy Corporation

1981 - 1985 Research Physicist, U.S. Army Ballistics Research Laboratory

Research Interests

Physical characteristics of wildland fuel combustion (energy transport, spectra, processes), spatial patterning of wildland fires, detection of fires using remote sensing methods, instrumentation development for wildfire and other environmental monitoring, remote sensing camera system development.

Relevant Publications and Published Presentations

R. Kremens, M. Dickinson, A. Bova, J. Faulring, Accurately mapping unburned areas using time-sequenced airborne imaging, presented at the Assoociation for Fire Ecology Mixed Severeity Fire Regime Conference, Spokane WA, November 2004

Donald McKeown, Jaun Cockburn, Jason Faulring, Robert Kremerns, David Morse, Harvey Rhody, Michael Richardson, Wildfire airborne sensor program (WASP): A new wildland fire detection and mapping system, published in the proceedings of the Tenth Biennial USDA Forest Service Remote Sensing Applications Conference, Salt Lake City, UT, April 5-9 2004

R. Kremens, J. Faulring, N. Raqueno, A Low-Cost Bathymeter using Commercial Off-The-Shelf Components, presented at the 46^th Annual International Association for Great Lakes Research, Chicago, Il, June 2003

Empirical Testing of Subpixel Detection of Fire, A.E. Ononye, A. Vodacek, R. Kremens, Y.Li and D. Merritt, presented at SPIE Aerosense 2003, Orlando Fl, April 2003

Low cost bathymetry loggers using off-the-shelf components, J. Faulring, R. Kremens, presented at the 2003 GLRC Conference, March 2003

Low Cost Autonomous Field-Deployable Environment Sensors, Robert L. Kremens, Andrew J. Gallagher, Adolph Seema, American Institute of Physics Proceedings of the Unattended Radiation Sensor Systems for Remote Applications, Vol 632, pp.190-199, April 15-17, 2002, Washington D.C.

Robert Kremens, Jason Faulring, Andrew Gallagher, A low cost weather/situation monitor for wildland firefighter safety, published in the proceedings of the International Association of Wildfire Safety Symposium, Toronto, CA, November 2003

Robert Kremens, Jason Faulring, Colin Hardy, Measurement of the time-temperature and emissivity history of the burn scar for remote sensing applications, published in the proceedings of the 2^nd Fire Ecology Congress (American Meteorological Society) Orlando Fl, November 2003

Kremens, R., J. Faulring, A. Gallagher, A. Seema, and A. Vodacek. Autonomous field- deployable wildland fire sensors. Int. J. Wildland Fire. 12, 237-244, 2002

Kremens, Robert L., Andrew J. Gallagher, Adolph Seema, Low Cost Autonomous Field - Deployable Environment Sensors, American Institute of Physics Proceedings of the Unattended Radiation Sensor Systems for Remote Applications Symposium, Vol. 632, July 2002.

Vodacek, A., R. L. Kremens, A. J. Fordham, S. C. VanGorden, D. Luisi, J. R. Schott, and D.J. Latham, Remote Optical Detection of Biomass Burning Using a Potassium Emission Signature International Journal of Remote Sensing 23, 2721-2726, 2002.

Kremens, R., A Semi-Automated Counting Method for Improved Accuracy in Invertebrate Whole Effluent Toxicity Tests, M.S. Thesis, Environmental Studies, University of Rochester, 2000.

Partial List of Other Refereed Publications (>30)

Marshall, F., J. Delletrez, V. Glebov, R.Town, B. Yaakobi, R. Kremens, M.Cable, Inertially Confined Plasmas, Dense Plasmas, Equations of State - Direct-drive, hollow-shell implosion studies on the 60-beam, UV OMEGA laser system. Physics of Plasmas, 7, 1006-1013, June 2000.

Sampat, N., Venkataraman, S., Yeh, T., Kremens, R.L., System implications of implementing auto-exposure on consumer digital camerasProc. SPIE Vol. 3650, p. 100-107, Sensors, Cameras, and Applications for Digital Photography, N. Sampat; T. Yeh, Eds., March 1999.

Knauer, J.P., Kremens, R. L., Russotto, M. A., and Tudman, S., Using cosmic rays to monitor large scintillator arrays, Rev. Sci. Inst., 66, 926-928, 1995.

Li, C., D. Hicks, R.Petrasso, F. Seguin, B. Burke, J. Knauer, S. Cremer, R. Kremens, M Cable, T. Phillips, Charged-Coupled Devices for Charged-Particle Spectroscopy on OMEGA and NOVA, Rev. Sci. Inst., 68, 593-595, January 1997.

Soures, J. M., McCrory, R. L., Boehly, T. R., Craxton, R. S., Jacobs, S. D., Kelly, J. H., Kessler, T. J., Knauer, J. P., Kremens, R. L., Kumpan, S. A., OMEGA upgrade laser for direct-drive target experiments, Laser and particle beams. 11, no. 2, 317-322, 1993.

Collaborators (alphabetical order): Joe Atkinson, NYU at Buffalo; Bob Brower, IAGT; Emily Constantine, IAGT; Colin Hardy, USFS; Jim Haynes, SUNY Brockport; Don Latham, USFS; Joe Makarewicz, SUNY Brockport; Bryce Nordgren, USFS; Dan Phillips, RIT; Lloyd Queen, U. of Montana; John Schott, RIT; Anthony Vodacek, RIT; Lisa Warnecke, GeoManagement Assoc.

WILLIAM (RUDDY) MELL

Education

Ph. D. 1994 Applied Mathematics, University of Washington, Seattle, WA

Dissertation: “An Investigation of Closure Models for Nonpremixed Turbulent Reacting Flow”

M.S. 1987 Applied Mathematics, University of Washington, Seattle, WA

B.S. 1981 Geophysics, University of Minnesota, Minneapolis, MN

Research Interests

Numerical simulation of combustion and large scale fire problems including fire spread through complex fuels, combustion in reduced gravity. See this site for more information: www2.bfrl.nist.gov/userpages/wmell/public.html.

Experience

-Research Scientist, National Institute of Standards and Technology, Gaithersburg, MD; 1994-1999; 2003-

Computer simulation of Wildland-Urban Interface and wildland fires, thermal radiation solvers, microgravity combustion, heat transfer in firefighter protective clothing.

-Research Professor, Mechanical Engineering Dept., University of Utah, Salt Lake City, UT; 1999-2003. Fire simulation, microgravity combustion, fire safety, numerical algorithms for heat transer problems.

-Research Assistant, Mechanical Engineering Dept., University of Washington, Seattle, WA; 1987-1994. Simulation and modeling of turbulent combustion.

-Research Assistant, Applied Physics Laboratory, University of Washington, Seattle; 1986. Modeling of shallow water seabed structure using inverse theory.

-Field Geophysicist, Dow Chemical, Crystal Falls, MI; 1983-1984. Mineral exploration.

Selected Refereed Journal Publications

1. W.E. Mell, K.B. McGrattan, H.R. Baum, “g-jitter Effects on Spherical Diffusion Flames,” Microgravity Science and Technology, 15, pp. 12-30 (2004)

2. W.E. Mell and J.R. Lawson, “A Heat Transfer Model for Fire Fighters’ Protective Clothing,” Fire Technology, 36, pp. 39-68 (2000)

3. W.E. Mell and T. Kashiwagi, “Dimensional Effects on the Transition from Ignition to Flame Spread in Microgravity,” Proc. Combust. Institute, 27, pp. 2635-2641 (1998).

4. H.R. Baum and W.E. Mell, “A Radiative Transport Model for Large Eddy Fire Simulations,” Combust. Theory and Modeling, 2, pp. 405-422 (1998).

5. W.E. Mell, K.B. McGrattan and H.R. Baum, “Numerical Simulation of Combustion in Fire Plumes,” Proc. Comb. Institute, Vol. 26, pp. 1523–1530 (1996).

6. W.E. Mell, V. Nilsen, G. Kosály and J.J. Riley, “Investigation of Closure Models for Nonpremixed Turbulent Reacting Flows,” Phys. of Fluids, 6, pp. 1331–1356 (1994).

Selected Recent Conference Presentations

1. R. Rehm, D. Evans, W. Mell, “Mathematical Modeling of Neighborhood-Scale Fires,” Symposium on the Role of Technology in Developing a Holistic Approach to Wildland-Urban Interface Fires, National Academy of Sciences, Organized by the U.S. General Accounting Office, Washington, D.C., August 17-18, 2004.

2. M.A. Jenkins, W.E. Mell, “Physics Based Computer Modeling of Fire Spread through Vegetative Fuels,” Third Annual Eastern Area Modeling Consortium Meeting, May 18-19, 2004, Lansing, MI.

3. R. Rehm, D. Evans, W. Mell, “Mathematical Modeling of WUI Fires,” invited talk presented at the May 19, 2004 workshop entitled, Fires at the Wildland-Urban Interface: What does the Future Hold?. Sponsored by the Risk Predicition Initiative (RPI) of the re-insurance industry at the Bermude Biological Station for Research, Bermuda, Dr. Rick Murnane, organizer.

4. R. Rehm, D. Evans, W. Mell, S. Hostikka, K. McGrattan, G. Forney, C. Bouldin, E. Baker “Neiborhood-Scale Fire Spread,” Second Intnl. Wildland Fire Ecology and Fire Managment Congress and Fifth Symposium on Fire and Forest Meteorology, American Meteorology Society, 16-20 November, 2003, Orland, FL

Awards

2001 Harry C. Bigglestone Award. NFPA Fire Protection Research Foundation

2001 Presidential Early Career Award for Scientists and Engineers

1994-1996 National Research Council —National Institue of Standards and Technology Postdoctoral Research Associateship

Professional Societies: American Physical Society, Combustion Institute, Sigma Xi Scientific Research Society

MARY ANN JENKINS

Mary Ann Jenkins, Associate Professor in the Department of Earth and Space Science and Engineering, York University, Toronto, Canada, and Adjunct Associate Professor in the Department of Meteorology, University of Utah, Salt Lake City, Utah. Currently Co-PI of ``Physically based wildland fire modeling and its integration in large-eddy atmospheric models,'' USDA Forest Service Agreement with University of Utah; in collaboration Dr. William Mell, National Institute of Standards and Technology; Graduate student Mr. Ruiyu Sun, Professor Steven K. Krueger, University of Utah,

Dept. of Meteorology; and Dr. Jay Charney, USDA Forest Service, Eastern Area Modeling Consortium. Research is focused on using a coupled wildfire--atmosphere numerical model to study the behavior of fires and fire plumes. Recent refereed publications on coupled fire-atmosphere modeling include:

Jenkins, M. A. 2002. An examination of the sensitivity of numerically simulated wildfires to low-level atmospheric stability and moisture and the consequences for the Haines Index. Int. J. of Wildland Fire 11(4):213-232.

Jenkins, M. A. 2004. Investigating the Haines Index using parcel model theory. Int. J. of Wildland Fire 13:1-13.

DONALD MCKEOWN

Mr. Donald McKeown recently joined Rochester Institute of Technology (RIT) as a Distinguished Researcher after 19 years experience with the Commercial and Government Systems Division at Eastman Kodak Company where he established a strong reputation in System Engineering, Program Management and Program Development in aerospace remote sensing technologies. He was Chief Payload Systems Engineer for the pioneering IKONOS high resolution commercial remote sensing camera system. He also managed a highly successful sensor technology development activity for next generation remote sensing systems. At RIT, McKeown is responsible for program development, program management, and system engineering for the Laboratory for Imaging Algorithms and Systems. Mr. McKeown is currently project manager on the 2002 NURI “Hyperspectral Algorithm Development and Prototype Exploitation Tool Demonstration Project” (NMA401-02-1-2004). He is also program manager and system engineer for the Wildfire Airborne Sensor Program (NASA NAG13-02051 and NAG13-03026) which is an airborne multispectral (Vis/IR) mapping camera system with a real time geospatial data processor.

ANTHONY BOVA

Anthony received a B.S. in physics from The Evergreen State College in Olympia, Washington. Working as a laboratory technician in the college’s scientific computing facility brought him into frequent contact with environmental scientists and kindled his interest in the application of physics to ecological systems. As an undergraduate he developed an original method, published in the International Journal of Fluid Dynamics, to visualize vortex formation over delta wings. His work with the Forest Service focuses on heat transfer, fluid dynamics and fire effects. Along with Dr. Matt Dickinson, he has an upcoming publication in the Canadian Journal of Forest Research on tree tissue necrosis resulting from fires, and manuscript in review on new methods for thermocouple use in fires. Anthony presented the preliminary results of Wildfire Airborne Sensor Project (WASP) data from a prescribed burn in SE Ohio at the 2004 meeting of the Ecological Society of America, and he and Dr. Dickinson are invited co-speakers at a fire ecology symposium on fire effects at the 2005 ESA meeting. Anthony’s most recent work includes coding subroutines that combine with mesoscale meteorological predictions to predict fuel moisture and fire intensities, and implementing inverse heat conduction algorithms to estimate surface heat flux of stems exposed to fire.

LETTERS OF SUPPORT

Letters are attached from the following cooperators:

REX MANN – Area Commander, Incident Command System, USFS

ELIZABETH REINHARDT – Scientist, Missoula Fire Science Laboratory, USFS

NEIL GIFFORD – Conservation Director, Albany Pine Bush Preserve Commission, TNC

KENNETH OUTCALT – Site Manager, Southern Coastal Plain Fire and Fire Surrogates Study Site, USFS

ROBERT MITCHELL – Scientist, Jones Ecological Research Center