Analysis of inventory data derived fuel characteristics and fire behavior under various environmental conditions.
Abstract not provided.
60 APPLIED LIFE SCIENCES↗
Engineering topics
Andreu, Anne
Publications and source records attributed to Andreu, Anne.
Effects of overstory composition and prescribed fire on fuel loading across a heterogeneous managed landscape in the southeastern USA.
In the southeastern USA, land use history, forest management and natural geomorphic features have created heterogeneous fuel loads. This apparent temporal and spatial variation in fuel loads make it difficult to reliably assess potential fire behavior from remotely sensed canopy variables to determine risk and to prescribe treatments. We examined this variation by exploring the relationships between overstory forest vegetation attributes, recent fire history, and selected surface fuel components across an 80,000 ha contiguous landscape. Measurements of dead and live vegetation components of surface fuels were obtained from 624 permanent plots, or about 1 plot per 100 ha of forest cover. Within forest vegetation groups, we modeled the relationship between individual surface fuel components and overstory stand age, basal area, site quality and recent fire history, then stochastically predicted fuel loads across the landscape using the same linkage variables. The fraction of the plot variation, i.e., R2, explained by predictive models for individual fuel components ranged from 0.05 to 0.66 for dead fuels and 0.03 to 0.97 for live fuels in pine dominated vegetation groups. Stand age and basal area were generally more important than recent fire history for predicting fuel loads. Mapped fuel loads using these regressor variables showed a very heterogeneous landscape even at the scale of a few square kilometers. The mapped patterns corresponded to stand based forest management disturbances that are reflected in age, basal area, and fire history. Recent fire history was significant in explaining variation in litter and duff biomass. Stand basal area was positively and consistently related to dead fuel biomass in most groups and was present in many predictive equations. Patterns in live fuel biomass were related to recent fire history, but the patterns were not consistent among forest vegetation groups. Age and basal area were related to live fuels in a complex manner that is likely confounded with periodic disturbances that disrupt stand dynamics. This study complements earlier hazardous fuels research in the southeastern USA, and indicates that succession, disturbance, site quality and decomposition interact with forest management practices to create variable spatial and temporal conditions. The inclusion of additional land use, disturbance history, and soil-topographic variables coupled to improved sampling methods may increase precision and subsequent fuel mapping.
09 BIOMASS FUELS↗
Developing custom fire behavior fuel models from ecologically complex fuel structures for upper Atlantic Coastal Plain forests.
Currently geospatial fire behavior analyses are performed with an array of fire behavior modeling systems such as FARSITE, FlamMap, and the Large Fire Simulation System. These systems currently require standard or customized surface fire behavior fuel models as inputs that are often assigned through remote sensing information. The ability to handle hundreds or thousands of measured surface fuelbeds representing the fine scale variation in fire behavior on the landscape is constrained in terms of creating compatible custom fire behavior fuel models. In this study, we demonstrate an objective method for taking ecologically complex fuelbeds from inventory observations and converting those into a set of custom fuel models that can be mapped to the original landscape. We use an original set of 629 fuel inventory plots measured on an 80,000 ha contiguous landscape in the upper Atlantic Coastal Plain of the southeastern United States. From models linking stand conditions to component fuel loads, we impute fuelbeds for over 6000 stands. These imputed fuelbeds were then converted to fire behavior parameters under extreme fuel moisture and wind conditions (97th percentile) using the fuel characteristic classification system (FCCS) to estimate surface fire rate of spread, surface fire flame length, shrub layer reaction intensity (heat load), non-woody layer reaction intensity, woody layer reaction intensity, and litter-lichen-moss layer reaction intensity. We performed hierarchical cluster analysis of the stands based on the values of the fire behavior parameters. The resulting 7 clusters were the basis for the development of 7 custom fire behavior fuel models from the cluster centroids that were calibrated against the FCCS point data for wind and fuel moisture. The latter process resulted in calibration against flame length as it was difficult to obtain a simultaneous calibration against both rate of spread and flame length. The clusters based on FCCS fire behavior parameters represent reasonably identifiable stand conditions, being: (1) pine dominated stands with more litter and down woody debriscomponents than other stands, (2) hardwood and pine stands with no shrubs, (3) hardwood dominated stands with low shrub and high non-woody biomass and high down woody debris, (4) stands with high grass and forb (i.e., non-woody) biomass as well as substantial shrub biomass, (5) stands with both high shrub and litter biomass, (6) pine-mixed hardwood stands with moderate litter biomass and low shrub biomass, and (7) baldcypress-tupelo stands. Models representing these stand clusters generated flame lengths from 0.6 to 2.3 musing a 30 km h{sub 1} wind speed and fireline intensities of 100-1500 kW m{sub 1} that are typical within the range of experience on this landscape. The fuel models ranked 1 < 2 < 7 < 5 < 4 < 3 < 6 in terms of both flame length and fireline intensity. The method allows for ecologically complex data to be utilized in order to create a landscape representative of measured fuel conditions and to create models that interface with geospatial fire models.
09 BIOMASS FUELS↗