Search

Abstract: In many forested ecosystems, it is increasingly recognized that the probability of burning is substantially reduced within the footprint of previously burned areas. This self‐limiting effect of wildland fire is considered a fundamental emergent property of ecosystems and is partly responsible for structuring landscape heterogeneity (i.e., mosaics of different age classes), thereby reducing the likelihood of uncharacteristically large fires in regions with active fire regimes. However, the strength and longevity of this self‐limiting phenomenon is not well understood in most fire‐prone ecosystems. In this study, we quantify the self‐limiting effect in terms of its strength and longevity for five fire‐prone study areas in western North America and investigate how each measure varies along a spatial climatic gradient and according to temporal (i.e., annual) climatic variation. Results indicate that the longevity (i.e., number of years) of the self‐limiting effect ranges between 15 yr in the warm and dry study area in the southwestern United States to 33 yr in the cold, northern study areas in located in northwestern Montana and the boreal forest of Canada. We also found that spatial climatic variation has a strong influence on wildland fire's self‐limiting capacity. Specifically, the self‐limiting effect within each study area was stronger and lasted longer in areas with low mean moisture deficit (i.e., wetter and cooler settings) compared to areas with high mean moisture deficit (warmer and drier settings). Last, our findings show that annual climatic variation influences wildland fire's self‐limiting effect: drought conditions weakened the strength and longevity of the self‐limiting effect in all study areas, albeit at varying magnitudes. Overall, our study provides support for the idea that wildland fire contributes to spatial heterogeneity in fuel ages that subsequently mediate future fire sizes and effects. However, our findings show that the strength and longevity of the self‐limiting effect varies considerably according to spatial and temporal climatic variation, providing land and fire managers relevant information for effective planning and management of fire and highlighting that fire itself is an important factor contributing to fire‐free intervals. [ABSTRACT FROM AUTHOR]

Southwestern ponderosa pine forests are vulnerable to fire‐driven conversion in a warming and drying climate, yet little is known about what kinds of ecological communities may replace them. To characterize postfire vegetation trajectories and their environmental determinants, plant assemblages (361 sample plots including 229 vascular plant species, surveyed in 2017) were sampled within eight burns that occurred between 2000 and 2003. I used nonmetric multidimensional scaling, k‐means clustering, principal component analysis, and random forest models to assess relationships between vegetation pattern, topographic and landscape factors, and gridded climate data. I describe seven postfire community types, including regenerating forests of ponderosa pine, aspen, and mixed conifers, shrub‐dominated communities of Gambel oak and mixed species, and herb‐dominated communities of native bunchgrasses and mixtures of ruderal, native, and nonnative species. Forest recovery was generally associated with cooler, mesic sites in proximity to forested refugia; shifts toward scrub and grassland types were most common in warmer, dryer locations distant from forested refugia. Under future climate scenarios, models project decreases in postfire forest recovery and increases in nonforest vegetation. However, forest to nonforest conversion was partially offset under a scenario of reduced burn severity and increased retention of forested refugia, highlighting important management opportunities. Burning trends in the southwestern United States suggest that postfire vegetation will occupy a growing landscape fraction, compelling renewed management focus on these areas and paradigm shifts that accommodate ecological change. I illustrate how management decisions around resisting, accepting, or directing change could be informed by an understanding of processes and patterns of postfire community variation and likely future trajectories. [ABSTRACT FROM AUTHOR]

The wildland–urban interface (WUI) is the focus of many important land management issues, such as wildfire, habitat fragmentation, invasive species, and human–wildlife conflicts. Wildfire is an especially critical issue, because housing growth in the WUI increases wildfire ignitions and the number of homes at risk. Identifying the WUI is important for assessing and mitigating impacts of development on wildlands and for protecting homes from natural hazards, but data on housing development for large areas are often coarse. We created new WUI maps for the conterminous United States based on 125 million individual building locations, offering higher spatial precision compared to existing maps based on U.S. census housing data. Building point locations were based on a building footprint data set from Microsoft. We classified WUI across the conterminous United States at 30‐m resolution using a circular neighborhood mapping algorithm with a variable radius to determine thresholds of housing density and vegetation cover. We used our maps to (1) determine the total area of the WUI and number of buildings included, (2) assess the sensitivity of WUI area included and spatial pattern of WUI maps to choice of neighborhood size, (3) assess regional differences between building‐based WUI maps and census‐based WUI maps, and (4) determine how building location accuracy affected WUI map accuracy. Our building‐based WUI maps identified 5.6%–18.8% of the conterminous United States as being in the WUI, with larger neighborhoods increasing WUI area but excluding isolated building clusters. Building‐based maps identified more WUI area relative to census‐based maps for all but the smallest neighborhoods, particularly in the north‐central states, and large differences were attributable to high numbers of non‐housing structures in rural areas. Overall WUI classification accuracy was 98.0%. For wildfire risk mapping and for general purposes, WUI maps based on the 500‐m neighborhood represent the original Federal Register definition of the WUI; these maps include clusters of buildings in and adjacent to wildlands and exclude remote, isolated buildings. Our approach for mapping the WUI offers flexibility and high spatial detail and can be widely applied to take advantage of the growing availability of high‐resolution building footprint data sets and classification methods. [ABSTRACT FROM AUTHOR]

Many forests in dry mountain regions are characterized by a lower elevational treeline. Understanding the controls on the position of lower treeline is important for predicting future forest distributional shifts in response to global environmental change. Lower treelines currently at their climate limit are expected to be more sensitive to changing climate, whereas lower treelines constrained by non‐climatic factors are less likely to respond directly to climate change but may be sensitive to other global change agents. In this study, we used existing vegetation classifications to map lower treelines for our 1.7 million km2 study region in the U. S. Intermountain West. We modeled topoclimatic drivers of lower treeline position for each of three dominant forest types to identify topoclimatically limited treelines. We then used spatial data of edaphic properties, recent fire, and land use to identify lower treelines potentially constrained above their ecophysiological limits by non‐climatic processes. We found that the lower treeline ecotone of pinyon–juniper woodlands is largely limited by topoclimate and is likely to be sensitive to increasing temperatures and associated droughts, though these effects may be heterogeneously distributed across the landscape. In contrast, dry mixed‐conifer lower treelines in the northern portion of the study area rarely reached their modeled topoclimatic limit, suggesting that non‐climatic processes, including fire and land use, constrain the lower treeline above its ecophysiological limits in this forest type. Our results suggest that much of the lower treeline in the Intermountain West is currently climate limited and will thus be sensitive to ongoing climate changes. Lower treelines in other arid or semi‐arid mountainous regions around the globe may also be strongly sensitive to climate, though treeline response to climate change will be mediated at the local scale by soil properties, biotic interactions, and natural or anthropogenic disturbances. Our regional study of lower treeline provides a framework for identifying the drivers of lower treeline formation and allows for more robust projections of future treeline dynamics, which are needed to anticipate shifting global distributions of the forest biome. [ABSTRACT FROM AUTHOR]

Extensive outbreaks of bark beetles have killed trees across millions of hectares of forests and woodlands in western North America. These outbreaks have led to spirited scientific, public, and policy debates about consequential increases in fire risk, especially in the wildland-urban interface ( WUI), where homes and communities are at particular risk from wildfires. At the same time, large wildfires have become more frequent across this region. Widespread expectations that outbreaks increase extent, severity, and/or frequency of wildfires are based partly on visible and dramatic changes in foliar moisture content and other fuel properties following outbreaks, as well as associated modeling projections. A competing explanation is that increasing wildfires are driven primarily by climatic extremes, which are becoming more common with climate change. However, the relative importance of bark beetle outbreaks vs. climate on fire occurrence has not been empirically examined across very large areas and remains poorly understood. The most extensive outbreaks of tree-killing insects across the western United States have been of mountain pine beetle ( MPB; Dendroctonus ponderosae), which have killed trees over >650,000 km2, mostly in forests dominated by lodgepole pine ( Pinus contorta). We show that outbreaks of MPB in lodgepole pine forests of the western United States have been less important than climatic variability for the occurrence of large fires over the past 29 years. In lodgepole pine forests in general, as well as those in the WUI, occurrence of large fires was determined primarily by current and antecedent high temperatures and low precipitation but was unaffected by preceding outbreaks. Trends of increasing co-occurrence of wildfires and outbreaks are due to a common climatic driver rather than interactions between these disturbances. Reducing wildfire risk hinges on addressing the underlying climatic drivers rather than treating beetle-affected forests. [ABSTRACT FROM AUTHOR]

Humans have a profound effect on fire regimes by increasing the frequency of ignitions. Although ignition is an integral component of understanding and predicting fire, to date fire models have not been able to isolate the ignition location, leading to inconsistent use of anthropogenic ignition proxies. Here, we identified fire ignitions from the Moderate Resolution Imaging Spectrometer ( MODIS) Burned Area Product (2000-2012) to create the first remotely sensed, consistently derived, and regionally comprehensive fire ignition data set for the western United States. We quantified the spatial relationships between several anthropogenic land-use/disturbance features and ignition for ecoregions within the study area and used hierarchical partitioning to test how the anthropogenic predictors of fire ignition vary among ecoregions. The degree to which anthropogenic features predicted ignition varied considerably by ecoregion, with the strongest relationships found in the Marine West Coast Forest and North American Desert ecoregions. Similarly, the contribution of individual anthropogenic predictors varied greatly among ecoregions. Railroad corridors and agricultural presence tended to be the most important predictors of anthropogenic ignition, while population density and roads were generally poor predictors. Although human population has often been used as a proxy for ignitions at global scales, it is less important at regional scales when more specific land uses (e.g., agriculture) can be identified. The variability of ignition predictors among ecoregions suggests that human activities have heterogeneous impacts in altering fire regimes within different vegetation types and geographies. [ABSTRACT FROM AUTHOR]

Widespread fire suppression and thinning have altered the structure and composition of many forests in the western United States, making them more susceptible to the synergy of large-scale drought and fire events. We examine how these changes affect carbon storage and stability compared to historic fire-adapted conditions. We modeled carbon dynamics under possible drought and fire conditions over a 300-year simulation period in two mixed-conifer conditions common in the western United States; (1) pine-dominated with an active fire regime and (2) fir-dominated, fire suppressed forests. Fir-dominated stands, with higher live- and dead-wood density, had much lower carbon stability as drought and fire frequency increased compared to pine-dominated forest. Carbon instability resulted from species (i.e., fir's greater susceptibility to drought and fire) and stand (i.e., high density of smaller trees) conditions that develop in the absence of active management. Our modeling suggests restoring historic species composition and active fire regimes can significantly increase carbon stability in fire-suppressed, mixed-conifer forests. Long-term management of forest carbon should consider the relative resilience of stand structure and composition to possible increases in disturbance frequency and intensity under changing climate. [ABSTRACT FROM AUTHOR]

National‐scale analyses of fire occurrence are needed to prioritize fire policy and management activities across the United States. However, the drivers of national‐scale patterns of fire occurrence are not well understood, and how the relative importance of human or biophysical factors varies across the country is unclear. Our research goal was to model the drivers of fire occurrence within ecoregions across the conterminous United States. We used generalized linear models to compare the relative influence of human, vegetation, climate, and topographic variables on fire occurrence in the United States, as measured by MODIS active fire detections collected between 2000 and 2006. We constructed models for all fires and for large fires only and generated predictive maps to quantify fire occurrence probabilities. Areas with high fire occurrence probabilities were widespread in the Southeast, and localized in the Mountain West, particularly in southern California, Arizona, and New Mexico. Probabilities for large‐fire occurrence were generally lower, but hot spots existed in the western and south‐central United States The probability of fire occurrence is a critical component of fire risk assessments, in addition to vegetation type, fire behavior, and the values at risk. Many of the hot spots we identified have extensive development in the wildland–urban interface and are near large metropolitan areas. Our results demonstrated that human variables were important predictors of both all fires and large fires and frequently exhibited nonlinear relationships. However, vegetation, climate, and topography were also significant variables in most ecoregions. If recent housing growth trends and fire occurrence patterns continue, these areas will continue to challenge policies and management efforts seeking to balance the risks generated by wildfires with the ecological benefits of fire. [ABSTRACT FROM AUTHOR]

Wildfires are becoming more frequent in parts of the globe, but predicting where and when wildfires occur remains difficult. To predict wildfire extremes across the contiguous United States, we integrate a 30‐yr wildfire record with meteorological and housing data in spatiotemporal Bayesian statistical models with spatially varying nonlinear effects. We compared different distributions for the number and sizes of large fires to generate a posterior predictive distribution based on finite sample maxima for extreme events (the largest fires over bounded spatiotemporal domains). A zero‐inflated negative binomial model for fire counts and a lognormal model for burned areas provided the best performance. This model attains 99% interval coverage for the number of fires and 93% coverage for fire sizes over a six year withheld data set. Dryness and air temperature strongly predict extreme wildfire probabilities. Housing density has a hump‐shaped relationship with fire occurrence, with more fires occurring at intermediate housing densities. Statistically, these drivers affect the chance of an extreme wildfire in two ways: by altering fire size distributions, and by altering fire frequency, which influences sampling from the tails of fire size distributions. We conclude that recent extremes should not be surprising, and that the contiguous United States may be on the verge of even larger wildfire extremes. [ABSTRACT FROM AUTHOR]