Robert E. Keane

Simulating Cumulative Fire Effects in Ponderosa Pine/Douglas-Fir Forests

A successional process model has been adapted for use with species from ponderosa pine/Douglas—fir (Pinus ponderosa var. ponderosa)/(Pseudotsuga menziesii var. glauca) forests of the inland Northwest. Its design allows modification for application to other forest types. This model, FIRSUM, simulates tree establishment, growth, and mortality, along with live and dead fuel accumulation, fire behavior, and fuel reduction on a 400—m2 plot. The modeling contains algorithms for influences on tree establishment and growth including temperature, water stress, light tolerance, and site quality. The model was used to predict 200 yr of forest succession for five different disturbance regimes. This allowed comparison of patterns of basal area by species, of duff and fuel accumulation, and on fire intensities among the following scenarios: (1) no fires (fire suppression), (2) consistent fire intervals of 10, 20, and 50 yr, and (3) a natural fire regime of variable intervals reconstructed from fire scarred trees. Frequent fires (10— and 20—yr intervals) were simulated to be of low intensity, resulting in scorch heights of 0.5—3.0 m. These fires prevented Douglas—fir saplings from surviving and becoming part of the overstory. Simulation results for the 10— and 20—yr fire intervals were similar to those for the natural fire regime, based on fire occurrence between AD 1600 and 1900. However, the occasional long fire intervals within the natural fire regime allowed greater regeneration success for ponderosa pine. Fires at regular intervals of 50 yr were more severe and resulted in a decrease of western larch (Larix occidentalis) after 150 yr, with a corresponding increase in ponderosa pine. Douglas—fir slowly increased in basal area and became established in the overstory after 200simulations years. The no—fire scenario allowed Douglas—fir to achieve dominance in the understory, and eventually in the overstory, thereby limiting survival of ponderosa pine and western larch regeneration. A test of the model showed predictions to be within 19% of field observations, and a sensitive analysis of FIRESUM showed parameters associated with the growth algorithm to be most critical for predicting successional trends.

Temperate and boreal forest mega-fires: characteristics and challenges

Mega-fires are often defined according to their size and intensity but are more accurately described by their socioeconomic impacts. Three factors – climate change, fire exclusion, and antecedent disturbance, collectively referred to as the “mega-fire triangle” – likely contribute to todays mega-fires. Some characteristics of mega-fires may emulate historical fire regimes and can therefore sustain healthy fire-prone ecosystems, but other attributes decrease ecosystem resiliency. A good example of a program that seeks to mitigate mega-fires is located in Western Australia, where prescribed burning reduces wildfire intensity while conserving ecosystems. Crown-fire-adapted ecosystems are likely at higher risk of frequent mega-fires as a result of climate change, as compared with other ecosystems once subject to frequent less severe fires. Fire and forest managers should recognize that mega-fires will be a part of future wildland fire regimes and should develop strategies to reduce their undesired impacts.

Mapping vegetation and fuels for fire management on the Gila National Forest Complex, New Mexico

Fuels and vegetation spatial data layers required by the spatially explicit fire growth model FARSITE were developed for all lands in and around the Gila National Forest in New Mexico. Satellite imagery, terrain modeling, and biophysical simulation were used to create the three vegetation spatial data layers of biophysical settings, cover type, and structural stage. Fire behavior fuel models and vegetation characteristics needed by FARSITE were assigned to combinations of categories on maps developed from sampled field data and also from estimates by local fire managers, ecologists, and resource specialists. FARSITE fuels maps will be used to simulate growth of fires on the Gila National Forest aiding managers in the planning and allocation of resources for managing fire. An extensive accuracy assessment of all maps indicated surface and crown fuels layers are about 30 to 40 percent accurate. This methodology was designed to be repli-cated for other areas of the western United States. The use of trade or firm names in this publication is for reader information and does not imply endorsement by the U.S. Department of Agriculture or any product or service Acknowledgments This project was a cooperative effort between the USDA Forest Service Rocky Mountain Research Station Fire Sciences Laboratory, the Gila National Forest, and the Fire Management and Engineering staffs of Region 3. We thank all who helped with the successful completion of this project including We would especially like to acknowledge the superior efforts of Owen Williams and Lawrence Garcia of the Gila National Forest for their help in field sampling and logistics. Abstract The Authors Robert E. Keane is a research ecologist with the USDA Forest Service, Rocky Mountain Re-has developed various ecological computer models for the Fire Effects Project to study both ecosystem management and research. His most recent research includes the mapping of fuels for fire growth and fire effects; synthesis of a First Order Fire Effects Model; construction of mechanis-tic ecosystem process models that integrate fire behavior and effects into succession simulation; restoration of whitebark pine in the Northern Rocky Mountains; and spatial simulation of succes-sional communities on the landscape using GIS and satellite imagery. He received his B.S. degree in forest engineering from the University of Maine, Orono; his M.S. degree in forest ecology from the University of Montana, Missoula; and his Ph.D. degree in forest ecology from the is currently working on GIS related projects for ecosystem management using remote …

Modelling stand dynamics in whitebark pine (Pinus albicaulis) forests

Abstract The ecological process model firesum (a fire su ccession mo del) has been adapted to whitebark pine ( Pinus albicaulis ) forests of the inland Northwest and Rocky Mountains, U.S.A. firesum simulates tree establishment, growth, and mortality on a 400-m 2 plot. Also modelled are live and dead fuel accumulations, fire behavior, fuel reduction, and insect and disease mortality. The following influences on tree establishment and growth are simulated in the model: temperature, water stress, site quality, and light conditions. An additional submodel in firesum simulates the mutualistic relationship between the Clarks nutcracker ( Nucifraga columbiana ) and the whitebark pine. Seed caches made by the Clarks nutcracker are evidently responsible for most whitebark pine regeneration. Whitebark pine seeds are also an important food for red squirrels ( Tamiasciurus hudsonicus ), black bear ( Ursus americana ), and especially grizzly bears ( Ursus arctos horribilis ). Infestations of white pine blister rust ( Cronartium ribicola ) and mountain pine beetle ( Dendroctonus ponderosae ), and successional replacement by shade-tolerant conifers can reduce whitebark pine populations, which can adversely impact dependent wildlife species. The model firesum was used to investigate the effects of fire, insects, and disease on whitebark pine regeneration and growth. Model predictions of basal area by tree species are presented for contrasting disturbance scenarios. Simulation predictions indicate severe reduction in some whitebark pine populations if current trends are not altered. Model predictions from a test were compared with inventory data from actual postfire stands.

The FireBGCv2 landscape fire and succession model: a research simulation platform for exploring fire and vegetation dynamics

Fire management faces important emergent issues in the coming years such as climate change, fire exclusion impacts, and wildland-urban development, so new, innovative means are needed to address these challenges. Field studies, while preferable and reliable, will be problematic because of the large time and space scales involved. Therefore, landscape simulation modeling will have more of a role in wildland fire management as field studies become untenable. This report details the design and algorithms of a complex, spatially explicit landscape fire and vegetation model called FireBGCv2. FireBGCv2 is a C++ computer program that incorporates several types of stand dynamics models into a landscape simulation platform. FireBGCv2 is intended as a research tool. Descriptions of FireBGCv2 code, sample input files, and sample output are included in this report, but this report is not intended as a users manual because the inherent complexity and wide scope of FireBGCv2 makes it unwieldy and difficult to use without extensive training. The primary purpose of this report is to document FireBGCv2 in adequate detail to interpret simulation results.

Forest succession on four habitat types in western Montana

Presents classifications of successional community types on four major forest habitat types in western Montana. Classifications show the sequences of seral community types developing after stand-replacing wildfire and clearcutting with broadcast burning, mechanical scarification, or no followup treatment. Information is provided for associating vegetational response to treatments.

Simulating historical landscape dynamics using the landscape fire succession model LANDSUM version 4.0

The range and variation of historical landscape dynamics could provide a useful reference for designing fuel treatments on todays landscapes. Simulation modeling is a vehicle that can be used to estimate the range of conditions experienced on historical landscapes. A landscape fire succession model called LANDSUMv4 (LANDscape SUccession Model version 4.0) is presented here as a tool for estimating historical range and variation (HRV) of landscape characteristics. The model simulates fire and succession on fine scale landscapes for land management applications. It simulates vegetation development as a deterministic process by changing the species composition and stand structure assigned to a polygon. Disturbance initiation is modeled stochastically and disturbance effects are based on the current vegetation conditions of the polygon. Details of all model algorithms are discussed and the model is demonstrated for two applications. Results of an extensive sensitivity and model behavior analysis are also presented.Please note: The data sets for this publication are only available on the CD. Ordering Information

Use of Expert Knowledge to Develop Fuel Maps for Wildland Fire Management

Fuel maps are becoming an essential tool in fire management because they describe, in a spatial context, the one factor that fire managers can control over many scales – surface and canopy fuel characteristics. Coarse-resolution fuel maps are useful in global, national, and regional fire danger assessments because they help fire managers effectively plan, allocate, and mobilize suppression resources (Burgan et al. 1998). Regional fuel maps are useful as inputs for simulating carbon dynamics, smoke scenarios, and biogeochemical cycles, as well as for describing fire hazards to support prioritization of firefighting resources (Leenhouts 1998; Lenihan et al. 1998).

Evaluating indices that measure departure of current landscape composition from historical conditions

A measure of the degree of departure of a landscape from its range of historical conditions can provide a means for prioritizing and planning areas for restoration treatments. There are few statistics or indices that provide a quantitative context for measuring departure across landscapes. This study evaluated a set of five similarity indices commonly used in vegetation community ecology (Sorensons Index, Chord Distance, Morisitas Index, Euclidean Distance, and Similarity Ratio) for application in estimating landscape departure (where departure = 1 - similarity). This involved comparing composition (vegetation type by area) of a set of reference landscapes to the compositions of 1,000 simulated historical landscapes. Stochastic simulation modeling was used to create a diverse set of synthetic reference and historical landscapes for departure index evaluation. Five reference landscapes were created to represent various degrees of expected departure from historical conditions. Both reference and historical landscapes were created to contain four important factors that could potentially influence departure calculation: (1) number of classes defining landscape composition, (2) dominance of the classes, (3) variability of area with the classes, and (4) temporal autocorrelation. We found that most evaluated indices are useful but not optimal for calculating departure. The Sorensons Index appeared to perform the best with consistent and precise behavior across the ranges of the four treatments. The number of classes used to describe vegetation had the strongest influence on index performance; landscape composition defined by few classes had the least accurate, most imprecise, and most highly variable departure estimates. While results from this study show the utility of similarity indices in evaluating departure, it is also evident that a new set of statistics are needed to provide a more comprehensive analysis of departure for future applications.

Mapping Ecological Attributes Using an Integrated Vegetation Classification System Approach

Summary Land managers need vegetation maps to inventory, monitor, and manage ecological resources across multiple spatial and temporal scales. Current vegetation maps usually only describe one vegetation characteristic, such as cover types, across the landscape. Although these maps provide important information for land management, they often fall short of addressing key issues like forest health and ecosystem management. In this paper we present an integrated approach where three different vegetation classifications are used in concert to spatially characterize many ecological attributes such as snag densities, insect susceptibility, and fire behavior across the landscape. Two examples from the Pacific Northwest are used to illustrate how this approach can be used to describe fuel characteristics and resource hazard across multiple scales.