Introduction

Many U.S. forests, especially those with historically short-interval low- to moderate-severity fire regimes, are too dense and have excessive quantities of fuels. Widespread treatments are needed to restore ecological integrity and reduce the high risk of destructive, uncharacteristically severe fires in these forests. Among possible treatments, however, the appropriate balance among cuttings, mechanical fuel treatments, and prescribed fire is often unclear. For improved decision-making, resource managers need much better information about the consequences of alternative management practices involving fire and mechanical/manual "fire surrogates."

Long-term, interdisciplinary research thus should be initiated to quantify the consequences and tradeoffs of alternative fire and fire surrogate treatments. Both ecological and economic aspects must be included as integral components. The research needs to be experimental, rather than retrospective or correlative, to permit stronger inferences about cause-and-effect relationships. Only through such research will it be possible to determine which ecosystem functions of fire can be emulated satisfactorily by other means, which may be irreplaceable, and the implications for management. The human dimensions of the problem are also important. Treatment costs and utilization economics, as well as social and political acceptability, strongly influence decisions about treatment alternatives. Such research must be a cooperative effort, involving land managers, researchers, and other interested parties.

A team of scientists and land managers has designed an integrated national network of long-term research sites to address this need, with support from the USDA/USDI Joint Fire Science Program. The steering committee and other participants in this national "Fire/Fire Surrogate" (FFS) study represent a number of federal and state agencies, universities, and private entities, as well as a wide range of disciplines and geographic regions. The study will use a common experimental design to facilitate broad applicability of results.

Objectives

Objectives of the project are as follows:

1. Quantify the initial effects (first five years) of fire and fire surrogate treatments on a number of specific core response variables within the general groupings of (a) vegetation, (b) fuel and fire behavior, (c) soils and forest floor (including relation to local hydrology), (d) wildlife, (e) entomology, (f) pathology, and (g) treatment costs and utilization economics.
2. Provide an overall research design that (a) establishes and maintains the study as an integrated national network of long-term interdisciplinary research sites utilizing a common "core" design to facilitate broad applicability of results, (b) allows each site to be independent for purposes of statistical analysis and modeling, as well as being a component of the national network, and (c) provides flexibility for investigators and other participants responsible for each research site to augment--without compromising--the core design as desired to address locally-important issues and to exploit expertise and other resources available to local sites.
3. Within the first five years of the study, establish cooperative relationships, identify and establish network research sites, collect baseline data, implement initial treatments, document treatment costs and short-term responses to treatments, report results, and designate FFS research sites as demonstration areas for technology transfer to professionals and for the education of students and the public.
4. Develop and maintain an integrated and spatially-referenced database format to be used to archive data for all network sites, facilitate the development of interdisciplinary and multi-scale models, and integrate results across the network.
5. Identify and field test, in concert with resource managers and users, a suite of response variables or measures that are: (a) sensitive to the fire and fire surrogate treatments, and (b) both technically and logistically feasible for widespread use in management contexts. This suite of measures will form much of the basis for management monitoring of operational treatments designed to restore ecological integrity and reduce wildfire hazard.
6. Over the life of the study, quantify the ecological and economic consequences of fire and fire surrogate treatments in a number of forest types and conditions in the United States. Develop and validate models of ecosystem structure and function, and successively refine recommendations for ecosystem management.

Research Approach

Experimental Design

The benefits of an integrated study with multiple experimental sites located around the country clearly can be enhanced if a common or "core" experimental design is utilized. The core experimental design for the FFS study--i.e., those elements of the design common to all research sites in the network--consists of common (1) treatments, (2) replication and plot size, and (3) response variables.

1. Treatments

The following suite of four FFS treatments will be implemented at each research site:

1. untreated control
2. prescribed fire only, with periodic reburns
3. initial and periodic cutting, each time followed by mechanical fuel treatment and/or physical removal of residue; no use of prescribed fire
4. initial and periodic cutting, each time followed by prescribed fire; fire alone also could be used one or more times between cutting intervals

These four treatments span a useful range both in terms of realistic management options and anticipated ecological effects. The non-control FFS treatments (treatments 2, 3, and 4) must be guided by a desired future condition (DFC) or target stand condition. The DFC will be defined mainly in terms of the tree component of the ecosystem--specifying such targets as diameter distribution, species composition, canopy closure, and spatial arrangements--and live and dead fuel characteristics. The following fire-related minimum standard will serve as a starting point for DFCs throughout the FFS network:

Each non-control treatment shall be designed to achieve stand and fuel conditions such that, if impacted by a head fire under 80th percentile weather conditions, at least 80 percent of the basal area of overstory (dominant and codominant) trees will survive. (See full proposal for further details.)

Research Approach - Continued

Given that this starting point is met for a given research site, however, the DFC can and should incorporate any additional management goals appropriate to the site and stand conditions and the expectations of resource managers and other stakeholders. Beyond the fire-related minimum standard for DFCs and the general treatment definitions given above, it is neither feasible nor desirable to prescribe detailed definitions of a core DFC or detailed treatment specifications that would apply across all research sites. Participants at each research site must provide this detail to ensure consistent application of treatments at that site.

2. Replication and Unit Size

Each treatment will be replicated at least 3 times at each research site, using either a completely randomized or randomized block design as appropriate to the research site. The core set of 4 treatments thus will be represented in 12 treatment units at a research site. Each of the 12 core treatment units at a research site will be 10-ha, within which core variables will be measured, surrounded by a buffer. The buffer, which is to be treated in the same way as the treatment unit it surrounds, will have a width at least equal to the height of a best site potential tree. Where feasible, the replicated units will be supplemented by much larger (200 to 400 ha or more), generally unreplicated areas treated to the same specifications, to facilitate the study of larger-scale ecological and economic/operational questions.

3. Response Variables

A major aspect of the common design proposed for this study is a set of core response variables to be measured at all the research sites. Core variables encompass several broad disciplinary areas, including vegetation, fuel and fire behavior, soils and forest floor, wildlife, entomology, pathology, and treatment costs and utilization economics. (A social science component probably will be linked to the study through no-cost cooperative arrangements and/or non-JFSP funding.) A corresponding set of disciplinary groups has had the responsibility for developing the core variables and associated measurement protocols, including coordinating across groups to ensure consistency, compatibility, and non-duplication of data collection efforts. Within-unit sampling of all variables will be keyed to a 50-m square grid of permanent sample points to be established and maintained in each treatment unit. Spatial referencing of all data to the grid will facilitate both spatial and cross-disciplinary analyses.

As suggested in Project Objective #2, the overall study is designed to balance the values of an integrated national network of research sites having a common design against the needs for each site to retain flexibility in addressing important local issues and in exploiting expertise and other resources available to that site. Accordingly, at the discretion of investigators, managers, and other participants involved in a given site, the core design may be augmented (provided it is not compromised) at that site by adding FFS treatments, adding one or more DFCs, adding replications, increasing treatment unit size (by increasing buffer width; the 10-ha treatment unit and core data collected within it would remain unchanged), and/or adding response variables. Except where additions to the core design are specifically justified for a given research site, we are requesting support through the Fire Science Program only for implementing the core design at each site.

Research Site Locations

In selecting research sites we developed and used the following set of criteria:

1. Site is representative of forests with a historically short-interval, low- to moderate-severity fire regime and a currently high risk of uncharacteristically severe fire.
2. Site is representative of widespread forest conditions (site characteristics, forest type and structure, treatment history) that are in need of, and likely to benefit from, fire or fire surrogate treatments, and in which such treatments are feasible.
3. Site contributes significantly to balancing the overall network in terms of regional representation and/or land ownership type.
4. Partners and cooperators are committed to and capable of participating in the program. This involves several factors, including: active support and interest in involvement on the part of partners/cooperators; available land base for the study; ability and willingness of land managers to implement the full suite of experimental treatments successfully within required time frame, repeat treatments over time as appropriate, commit selected sites for long-term research uses, and document these commitments in amendments to long-term land management plans.
5. On federal lands, treatment costs are borne by lead agency or partner.
6. Partnerships exist across agencies and with universities, and between researchers and managers.

The proposed initial network comprises 10 main sites and 1 satellite site (satellite will have less than the full suite of core treatments):

1. Mission Creek, north-central Washington, Wenatchee National Forest.
2. Hungry Bob, Blue Mountains of northeast Oregon, Wallowa-Whitman National Forest.
3. Lubrecht Forest, University of Montana, northern Rockies, western Montana.
4. Klamath Mountains, northwestern California, one or more national forests, possibly other ownerships.
5. Blodgett Forest Research Station, University of California-Berkeley, central Sierra Nevada, California.
6. Sequoia National Park, southern Sierra Nevada, California (satellite to Blodgett Forest Research Station site).
7. Southwest Plateau, Coconino and Kaibab National Forests, northern Arizona.
8. Jemez Mountains, Santa Fe National Forest, northern New Mexico.
9. Ohio Hill Country, lands managed by the Wayne National Forest, the Ohio Division of Forestry, Mead Paper Corporation, and The Nature Conservancy, southern Ohio.
10. Southeastern Piedmont, Clemson Experimental Forest, northwestern South Carolina.
11. Florida Coastal Plain, Myakka River State Park, southwest Florida.

All of these initial sites represent forests with a historically short-interval, low- to moderate-severity fire regime. Eight sites are in western coniferous forests, ranging from the Pacific Northwest to the Southwest. These sites share the fact that ponderosa pine is an important tree component, but sites vary in composition of other conifers and differ substantially in topographic and soil parameters. Two sites are in the southeastern U.S.--one in the Piedmont and one on the Coastal Plain--and are dominated by mixtures of southern pines with hardwood understories. Rounding out the network is a site in the midwestern oak-hickory type of Ohio. Collectively, these sites comprise a network that is truly national in scope. Depending on the level of interest and support available, future sites in the same or other fire regimes may be added to the network.