Article

Characterizing past fire occurrence in longleaf pine ecosystems with the Mid-Infrared Burn Index and a Random Forest classifier

Authors:
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

Prior to European settlement the longleaf pine (Pinus palustris) ecosystem covered over 92 million hectares in the southeastern United States. Historically, fire was an important driver of species composition in the longleaf pine ecosystem, but fire exclusion since the early 20th century has led to the degradation of longleaf pine communities and has had a detrimental effect on the large number of rare and endemic species found within this system. Thus, accurate estimates of fire history are important for better informed management of longleaf pine communities. Recently, satellite imagery has been used to identify burned areas. However, results have been inconsistent across physiographic regions and vegetation types (e.g. wetlands under high canopy). We developed a model using Landsat satellite imagery, coupled with a Random Forest (RF) machine learning algorithm, to identify burned areas and estimate the fire history from 1991 to 2019 for Fort Bragg, NC, one of the largest contiguous areas of longleaf pine ecosystem remaining. We calculated six spectral indices from the Landsat band values, including the Mid-Infrared Burn Index (MIRBI) and the change in MIRBI through time (ΔMIRBI), and used them as predictors in our RF model. We used the developed RF model to estimate the fire history for all known populations of 24 rare upland and wetland plant species found on Fort Bragg. We compared our results to a recent continental U.S. fire occurrence dataset, as well as the prescribed fire records from Fort Bragg. The overall AUC (area under the curve) for our RF model (0.74) compared favorably to the continental U.S. dataset results for Fort Bragg (0.69), and was able to capture the reduced fire frequency in wetlands. The most important predictor in our RF model was ΔMIRBI. Depending on the model, individual plant species were estimated to have experienced significant differences in fire frequency relative to the prescribed fire records. For our RF model, we estimated that 50% of wetland and 25% of upland species experienced a lower fire frequency relative to that represented in the prescribed fire records. The burn probability and classification tool generated in this paper provides land managers in the southeastern U.S. with a novel approach for accurately identifying burned areas and estimating local fire frequency across landscapes.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... In [145], random forest also is used. Satellite imagery is used in [146], coupled with a random forest algorithm, both to detect recently burned areas and also to estimate the fire history. ...
Article
Full-text available
Wildfires threaten and kill people, destroy urban and rural property, degrade air quality, ravage forest ecosystems, and contribute to global warming. Wildfire management decision support models are thus important for avoiding or mitigating the effects of these events. In this context, this paper aims at providing a review of recent applications of machine learning methods for wildfire management decision support. The emphasis is on providing a summary of these applications with a classification according to the case study type, machine learning method, case study location, and performance metrics. The review considers documents published in the last four years, using a sample of 135 documents (review articles and research articles). It is concluded that the adoption of machine learning methods may contribute to enhancing support in different fire management phases.
Article
Full-text available
Understanding demographic vital rates and the factors that affect those rates are key components of successful conservation strategies for many threatened and endangered rare plant species. Lindera subcoriacea is a rare dioecious shrub that occupies isolated wetland habitats in a small number of locations in the southeastern United States. The species faces a number of threats to its continued persistence, including habitat destruction, invasive species, and population isolation. From 2011 to 2019, we collected demographic information from 290 L. subcoriacea individuals within 28 populations on Fort Bragg, North Carolina and used the data to estimate demographic vital rates in unburned populations and after being exposed to prescribed fire. We then constructed population matrices and estimated population growth rates under a 3-, 5-, and 10-year return interval. Results indicated that L. subcoriacea individuals have high survivorship in both burned and unburned populations, seed production was reduced 1- and 2-year post-fire, seed production was highly uneven across individuals, seedling recruitment was extremely low, and simulated population growth rates were only above 1.0 under the 10-year fire return interval. Taken together, these results indicate that (1) L. subcoriacea populations are persisting with population growth rates close to one, (2) the short-term impacts of fire on the overall population growth rate of L. subcoriacea, while only 2–3% may determine long-term population viability, and (3) extremely uneven seed production and limited recruitment of seedlings into larger size classes make L. subcoriacea populations vulnerable to stochastic demographic processes.
Article
Full-text available
In frequent‐fire forests, wildland fire acts as a self‐regulating process creating forest structures that consist of a fine‐grained mosaic of isolated trees, tree groups of various sizes, and non‐treed openings. Though the self‐regulation of forest structure through repeated fires is acknowledged, few studies have investigated the role that fine‐scale pattern‐process linkages play in determining fire behavior and effects. To better understand the physical mechanisms driving these pattern‐process linkages, we used a three‐dimensional, physics‐based fire behavior model to investigate how the local arrangement of canopy fuels influences heat transfer from a surface fire to tree crowns and subsequent crown ignition and consumption. In particular, we were interested in the impacts of tree group size and crown separation distance on heat transfer. We found increased convective cooling for isolated individual trees and 3‐tree groups as compared to larger 7‐ and 19‐tree groups which resulted in a reduction of the net energy transferred from the surface fire to the tree crowns. Because isolated individuals and 3‐tree groups are exposed to less thermal energy, they require a greater surface fireline intensity to initiate torching and have less crown consumption than trees within larger groups. Similarly, we found that increased crown separation distance also reduced heat transfer and crown ignition. However, differences in crown ignition and consumption among various sized groups and separation distances depended upon the surface fireline intensity, suggesting that any change in crown consumption or tree mortality due to pattern‐process linkages may be best viewed as a conditional in nature. These findings identify the potential physical mechanisms responsible for supporting the complex forest structures typical of high‐frequency fire regimes, and the results may be useful for managers designing fuel hazard reduction and ecological restoration treatments.
Article
Full-text available
Complete and accurate burned area map data are needed to document spatial and temporal patterns of fires, to quantify their drivers, and to assess the impacts on human and natural systems. In this study, we developed the Landsat Burned Area (BA) algorithm, an update from the Landsat Burned Area Essential Climate Variable (BAECV) algorithm. Here, we present the BA algorithm and products, changes relative to the BAECV algorithm and products, and updated validation metrics. We also present spatial and temporal patterns of burned area across the conterminous U.S., how burned area varies in relation to the number of operational Landsat sensors, and a comparison with other burned area datasets, including the BAECV, Monitoring Trends in Burn Severity (MTBS), GeoMAC, and Moderate Resolution Imaging Spectroradiometer (MODIS) MCD64A1.006 data. The BA algorithm identifies burned areas in analysis ready data (ARD) time-series of Landsat imagery from 1984 through 2018 using machine learning, thresholding, and image segmentation. Validation with reference data from high-resolution commercial satellite imagery resulted in omission and commission error rates averaging 19% and 41%, respectively. In comparison, validation with Landsat reference data had omission and commission error rates averaging 40% and 28%, respectively when burned areas in cultivated crops and pasture/hay land-cover types were excluded. Both validation tests documented lower commission error rates relative to the BAECV products. The amount of burned area detected varies not only in response to climate but also with the number of operational sensors and scenes collected. The combined amount of burned area detected by multiple sensors was larger than from any individual sensor, but there was no significant difference between individual sensors. Therefore, we used BA products from individual sensors to assess trends over time and all available sensors to compare with other existing BA products. From 1984 through 2018, annual burned area averaged 30,000 km2, ranged between 14,000 km2 in 1991 and 46,500 km2 in 2012, and increased over time at a rate of 356 km2/year. Compared to existing burned area products, the new Landsat BA products identified 29% more burned area than the BAECV products (1984–2015), 183% more than the MTBS/GeoMAC products (1984–2018), and 56% more than the MCD64A1.006 products (2003–2018). The products had similar patterns of year-to-year variability; the R2 values of linear regressions between annual burned area were >0.70 with the BAECV products and the MTBS/GeoMAC products, but somewhat lower for the MCD64A1.006 product (R2 = 0.66). The BA products are routinely produced as new Landsat data are collected and provide a unique data source to monitor and assess the spatial and temporal patterns and the impacts of fire.
Article
Full-text available
Tropical savannas have been increasingly viewed as an opportunity for carbon sequestration through fire suppression and afforestation, but insufficient attention has been given to the consequences for biodiversity. To evaluate the biodiversity costs of increasing carbon sequestration, we quantified changes in ecosystem carbon stocks and the associated changes in communities of plants and ants resulting from fire suppression in savannas of the Brazilian Cerrado, a global biodiversity hotspot. Fire suppression resulted in increased carbon stocks of 1.2 Mg ha⁻¹ year⁻¹ since 1986 but was associated with acute species loss. In sites fully encroached by forest, plant species richness declined by 27%, and ant richness declined by 35%. Richness of savanna specialists, the species most at risk of local extinction due to forest encroachment, declined by 67% for plants and 86% for ants. This loss highlights the important role of fire in maintaining biodiversity in tropical savannas, a role that is not reflected in current policies of fire suppression throughout the Brazilian Cerrado. In tropical grasslands and savannas throughout the tropics, carbon mitigation programs that promote forest cover cannot be assumed to provide net benefits for conservation.
Article
Full-text available
Complete and accurate burned area data are needed to document patterns of fires, to quantify relationships between the patterns and drivers of fire occurrence, and to assess the impacts of fires on human and natural systems. Unfortunately, in many areas existing fire occurrence datasets are known to be incomplete. Consequently, the need to systematically collect burned area information has been recognized by the United Nations Framework Convention on Climate Change and the Intergovernmental Panel on Climate Change, which have both called for the production of essential climate variables (ECVs), including information about burned area. In this paper, we present an algorithm that identifies burned areas in dense time-series of Landsat data to produce the Landsat Burned Area Essential Climate Variable (BAECV) products. The algorithm uses gradient boosted regression models to generate burn probability surfaces using band values and spectral indices from individual Landsat scenes, lagged reference conditions, and change metrics between the scene and reference predictors. Burn classifications are generated from the burn probability surfaces using pixel-level thresholding in combination with a region growing process. The algorithm can be applied anywhere Landsat and training data are available. For this study, BAECV products were generated for the conterminous United States from 1984 through 2015. These products consist of pixel-level burn probabilities for each Landsat scene, in addition to, annual composites including: the maximum burn probability and a burn classification. We compared the BAECV burn classification products to the existing Global Fire Emissions Database (GFED; 1997–2015) and Monitoring Trends in Burn Severity (MTBS; 1984–2013) data. We found that the BAECV products mapped 36% more burned area than the GFED and 116% more burned area than MTBS. Differences between the BAECV products and the GFED were especially high in the West and East where the BAECV products mapped 32% and 88% more burned area, respectively. However, the BAECV products found less burned area than the GFED in regions with frequent agricultural fires. Compared to the MTBS data, the BAECV products identified 31% more burned area in the West, 312% more in the Great Plains, and 233% more in the East. Most pixels in the MTBS data were detected by the BAECV, regardless of burn severity. The BAECV products document patterns of fire similar to those in the GFED but also showed patterns of fire that are not well characterized by the existing MTBS data. We anticipate the BAECV products will be useful to studies that seek to understand past patterns of fire occurrence, the drivers that created them, and the impacts fires have on natural and human systems.
Article
Full-text available
Google Earth Engine is a cloud-based platform for planetary-scale geospatial analysis that brings Google's massive computational capabilities to bear on a variety of high-impact societal issues including deforestation, drought, disaster, disease, food security, water management, climate monitoring and environmental protection. It is unique in the field as an integrated platform designed to empower not only traditional remote sensing scientists, but also a much wider audience that lacks the technical capacity needed to utilize traditional supercomputers or large-scale commodity cloud computing resources.
Article
Full-text available
The Landsat Burned Area Essential Climate Variable (BAECV), developed by the U.S. Geological Survey (USGS), capitalizes on the long temporal availability of Landsat imagery to identify burned areas across the conterminous United States (CONUS) (1984–2015). Adequate validation of such products is critical for their proper usage and interpretation. Validation of coarse-resolution products often relies on independent data derived from moderate-resolution sensors (e.g., Landsat). Validation of Landsat products, in turn, is challenging because there is no corresponding source of high-resolution, multispectral imagery that has been systematically collected in space and time over the entire temporal extent of the Landsat archive. Because of this, comparison between high-resolution images and Landsat science products can help increase user's confidence in the Landsat science products, but may not, alone, be adequate. In this paper, we demonstrate an approach to systematically validate the Landsat-derived BAECV product. Burned area extent was mapped for Landsat image pairs using a manually trained semi-automated algorithm that was manually edited across 28 path/rows and five different years (1988, 1993, 1998, 2003, 2008). Three datasets were independently developed by three analysts and the datasets were integrated on a pixel by pixel basis in which at least one to all three analysts were required to agree a pixel was burned. We found that errors within our Landsat reference dataset could be minimized by using the rendition of the dataset in which pixels were mapped as burned if at least two of the three analysts agreed. BAECV errors of omission and commission for the detection of burned pixels averaged 42% and 33%, respectively for CONUS across all five validation years. Errors of omission and commission were lowest across the western CONUS, for example in the shrub and scrublands of the Arid West (31% and 24%, respectively), and highest in the grasslands and agricultural lands of the Great Plains in central CONUS (62% and 57%, respectively). The BAECV product detected most (> 65%) fire events > 10 ha across the western CONUS (Arid and Mountain West ecoregions). Our approach and results demonstrate that a thorough validation of Landsat science products can be completed with independent Landsat-derived reference data, but could be strengthened by the use of complementary sources of high-resolution data.
Article
Full-text available
Tropical savannas are a globally extensive biome prone to rapid vegetation change in response to changing environmental conditions. Via a meta-analysis, we quantified savanna woody vegetation change spanning the last century. We found a global trend of woody encroachment that was established prior the 1980s. However, there is critical regional variation in the magnitude of encroachment. Woody cover is increasing most rapidly in the remaining uncleared savannas of South America, most likely due to fire suppression and land fragmentation. In contrast, Australia has experienced low rates of encroachment. When accounting for land-use, African savannas have a mean rate annual woody cover increase two and a half times that of Australian savannas. In Africa, encroachment occurs across multiple land uses and is accelerating over time. In Africa and Australia, rising atmospheric CO2 , changing land management and rainfall are likely causes. We argue that the functional traits of each woody flora, specifically the N-fixing ability and architecture of woody plants, are critical to predicting encroachment over the next century and that African savannas are at high risk of widespread vegetation change. This article is protected by copyright. All rights reserved.
Article
Full-text available
Increasing awareness of the issue of deforestation and degradation in the tropics has resulted in efforts to monitor forest resources in tropical countries. Advances in satellite-based remote sensing and ground-based technologies have allowed for monitoring of forests with high spatial, temporal and thematic detail. Despite these advances, there is a need to engage communities in monitoring activities and include these stakeholders in national forest monitoring systems. In this study, we analyzed activity data (deforestation and forest degradation) collected by local forest experts over a 3-year period in an Afro-montane forest area in southwestern Ethiopia and corresponding Landsat Time Series (LTS). Local expert data included forest change attributes, geo-location and photo evidence recorded using mobile phones with integrated GPS and photo capabilities. We also assembled LTS using all available data from all spectral bands and a suite of additional indices and temporal metrics based on time series trajectory analysis. We predicted deforestation, degradation or stable forests using random forest models trained with data from local experts and LTS spectral-temporal metrics as model covariates. Resulting models predicted deforestation and degradation with an out of bag (OOB) error estimate of 29% overall, and 26% and 31% for the deforestation and degradation classes, respectively. By dividing the local expert data into training and operational phases corresponding to local monitoring activities, we found that forest change models improved as more local expert data were used. Finally, we produced maps of deforestation and degradation using the most important spectral bands. The results in this study represent some of the first to combine local expert based forest change data and dense LTS, demonstrating the complementary value of both continuous data streams. Our results underpin the utility of both datasets and provide a useful foundation for integrated forest monitoring systems relying on data streams from diverse sources.
Article
Full-text available
Positive feedbacks influenced by direct and indirect interactions between fire, vegetation, and microclimate can allow pyrophilic and pyrophobic ecosystems to co-occur in the same landscape, resulting in the juxtaposition of flammable and non-flammable vegetation. To quantify the drivers of these feedbacks, we combined measurements of vegetation, fuels, and microclimate with observations of fire spread along ecotonal gradients. We established 113 permanent transects (consisting of 532 plots), each traversing an ecotone between savanna and wetland in the Sandhills of North Carolina, USA. In each plot, we recorded cover of ten plant functional types. We collected surface fuels at a subset of our transects. We continuously monitored microclimate (nine meteorological variables) across 21 representative ecotones. Following prescribed fire, we measured fire spread along each transect. Vegetation structure and microclimate significantly predicted fire spread along the savanna-wetland ecotone. Fire spread was most influenced by vegetation structure, specifically C-4 grass cover, which accounted for 67 % of the variance explained by our model. We have identified the components of the fire, vegetation, and microclimate feedback that control where fires stop under current conditions, but their control should not be considered absolute. For example, when ignited in savanna, prescribed burns continued through wetland vegetation 43 % of the time. The feedback operating within these systems may be relatively weak as compared to other savanna systems. Environmental changes may alter fire spread extent, and with it ecosystem boundaries, or even ecosystem states.
Article
Full-text available
Scientific models that guide restoration/management protocols should be reviewed periodically as new data become available. We examine ecological concepts used to guide restoration of pine savannas and woodlands, historically prominent but now rare habitats in the southern North American Coastal Plain. For many decades, pine savanna management has been guided predominantly by a biome-centric succession model. Pine savannas have been considered early-successional communities that, in the absence of fire, transition rapidly toward closed-canopy hardwood forests. Recurrent fires have been viewed as exogenous disturbances that maintain savanna ecosystems as a sub-climax, blocking succession to an equilibrium steady state (closed-canopy forests). Over recent decades, a vegetation-fire feedback model has emerged in which pine savannas are conceptualized as persistent, non-equilibrium communities maintained by endogenous, co-evolutionary vegetation-fire feedbacks. Endemic plant species are resistant to fires and specialized for post-fire conditions generated by frequent lightning fires, primarily within a distinct fire season. These species produce pyrogenic fine fuels that are easily ignited. The resulting fire regimes, entrained by these vegetation-fire feedbacks, are predicted to result in persistent pine savannas. Local variation over space and time in evolutionary feedback mechanisms between pyrogenic vegetation and fire regimes produces heterogeneous landscapes. Disturbances of these feedbacks, such as human fire suppression, are postulated to result in rapid transition to communities lacking feedback elements, such as closed-canopy forest and those without pyrogenic species. Succession-based management focuses on reversing the transition to forest, primarily by removing hardwoods and reintroducing fire as a disturbance. However, we advocate restoration and management approaches that target reinstitution of functional vegetation-fire feedbacks. Such approaches should favor native pyrogenic plant species and reinstitute fire regimes that mimic historical, evolutionarily derived fire regimes. Vegetation-fire feedback concepts should be useful in addressing resistance and resilience of fiery ecosystems worldwide to inherent changes in feedback mechanisms, constituting a framework useful in addressing global management challenges.
Article
Full-text available
Southeastern United States habitats dominated by longleaf pine (Pinus palustris P. Miller) have declined precipitously in area and extent. Conservation of diverse ground-layer vegetation in these endangered habitats depends on prescribed fire. While the need for prescribed fire is now generally accepted, there is disagreement concerning the most appropriate fire regime. One of the more important variables is frequency of fire. Several hypothetical relationships between fire frequency and vascular plant richness and composition are suggested by the existing literature. Results of two long-term prescribed fire studies support the hypothesis that burning as frequently as fuels permit is optimal for maintaining the largest number of native ground-layer plant species. However, fire frequency effects on species composition differed between the two studies. Increasing fire frequency in South Carolina Ultisol flatwoods and wet savannas was associated with a distinct shift from woody to herbaceous-dominated communities. Herbs, particularly bunchgrasses and perennial forbs, dominated annual- and biennial-burn treatment plots, whereas triennial- and quadrennial-burn plots were shrub-dominated. In contrast, annual and biennial fires did not produce herbaceous dominated ground-layer vegetation in North Florida Spodosol flatwoods. Reduced dominance of saw palmetto and somewhat increased importance of forbs and grasses, particularly rhizomatous grasses, distinguished the annually burned plots. However, biennial- and quadrennial-burn plots were similar in composition and did not differ significantly in species richness at the largest spatial scale.
Article
Full-text available
Whether ignited by lightning or by Native Americans, fire once shaped many North American ecosystems. Euro-American settlement and 20th-century fire suppression practices drastically altered historic fire regimes, leading to excessive fuel accumulation and uncharacteristically severe wildfires in some areas and diminished flammability resulting from shifts to more fire-sensitive forest species in others. Prescribed fire is a valuable tool for fuel management and ecosystem restoration, but the practice is fraught with controversy and uncertainty. Here, we summarize fire use in the forests and woodlands of North America and the current state of the practice, and explore challenges associated with the use of prescribed fire. Although new scientific knowledge has reduced barriers to prescribed burning, societal aversion to risk often trumps known, long-term ecological benefits. Broader implementation of prescribed burning and strategic management of wildfires in fire-dependent ecosystems will require improved integration of science, policy, and management, and greater societal acceptance through education and public involvement in land-management issues.
Article
Full-text available
Although fire is a common disturbance in shrub-steppe, few studies have specifically tested burned area mapping accuracy in these semiarid to arid environments. We conducted a preliminary assessment of the accuracy of the Monitoring Trends in Burn Severity (MTBS) burned area product on four shrub-steppe fires that exhibited varying degrees of within-fire patch heterogeneity. Independent burned area perimeters were derived through visual interpretation and were used to cross-compare the MTBS burned area perimeters with classifications produced using set thresholds on the Relativised differenced Normalised Burn Index (RdNBR), Mid-infrared Burn Index (MIRBI) and Char Soil Index (CSI). Overall, CSI provided the most consistent accuracies (96.3-98.6%), with only small commission errors (1.5-4.4%). MIRBI also had relatively high accuracies (92.2-97.9%) and small commission errors (2.1-10.8%). The MTBS burned area product had higher commission errors (4.3-15.5%), primarily due to inclusion of unburned islands and fingers within the fire perimeter. The RdNBR burned area maps exhibited lower accuracies (92.9-96.0%). However, the different indices when constrained by the MTBS perimeter provided variable results, with CSI providing the highest and least variable accuracies (97.4-99.1%). Studies seeking to use MTBS perimeters to analyse trends in burned area should apply spectral indices to constrain the final burned area maps. The present paper replaces a former paper of the same title (http://dx.doi.org/10.1071/WF13206), which was withdrawn owing to errors discovered in data analysis after the paper was accepted for publication.
Article
Full-text available
Papers on the phytogeography of the North American coastal plain are poorly represented in the botanical literature, despite the fact that the coastal plain is one of the continent's major floristic provinces. Some authors have discussed plant distribution and/or endemism for local regions, but none have treated the coastal plain as a whole. We synthesized data from many sources and have compiled lists of all endemic families, genera, species, and infraspecific taxa. Analysis of these lists, and of county-level range maps, yields recurring patterns of species distributions. We define and map 27 such patterns and list representative taxa for each. The boundaries of many of these patterns appear to circumscribe centers of endemism. We discuss underlying reasons for the high diversity and endemism in the coastal plain, as well as the geographical derivation of the endemics, the age of the flora, and the role of past geological events.
Article
Full-text available
Fire-prone savanna ecosystems in southern African conservation areas are managed by prescribed burning in order to conserve biodiversity. A prescribed burning system designed to maximise the benefits of a diverse fire regime in savanna conservation areas is described. The area burnt per year is a function of the grass fuel load, and the number of fires per year is a function of the percentage area burnt. Fires are point-ignited, under a range of fuel and weather conditions, and allowed to burn out by themselves. The seasonal distribution of planned fires over a year is dependent on the number of fires. Early dry season fires (May–June) tend to be small because fuels have not yet fully cured, while late season fires (August–November) are larger. More fires are ignited in the early dry season, with fewer in the late dry season. The seasonality, area burnt, and fire intensity are spatially and temporally varied across a landscape. This should result in the creation of mosaics, which should vary in extent and existence in time. Envelopes for the accumulated percentage to be burnt per month, over the specified fire season, together with upper and lower buffers to the target area are proposed. The system was formalised after 8 years of development and testing in Pilanesberg National Park, South Africa. The spatial heterogeneity of fire patterns increased over the latter years of implementation. This fire management system is recommended for savanna conservation areas of >20 000 ha in size.
Article
Full-text available
Remote sensing can quantify past and present fire activity at spatial scales useful for a range of fire and vegetation management applications. In this study, we present a new automated approach to classifying burnt areas across the state of Queensland, Australia. The method is applied to complete time series of Landsat TM/ETM + imagery rather than single images and considers spectral (band 4, B4, and bands 4 + 5, B45), thermal, temporal and contextual information within a hierarchical framework. To maximise the available observations and the burnt area detected, we used imagery containing up to 60% cloud that was screened during pre-processing. Median filters were applied to smooth the time series and multi-date change detection used to locate negative outliers (large declines in reflectance relative to the median-smoothed time series). Watershed region growing was used to segment and map a larger spatial extent of the change while minimising commission errors. These segmented change objects were attributed as either burnt or unburnt using their thermal, reflective and contextual characteristics in a classification tree. Thermal information was found to be more important than reflective indices in the change attribution. Algorithm calibration used training data from ten Path/Rows located strategically across Queensland with four images sampled per path row (n = 40). Thresholds were optimised to maximise the burnt area detected while limiting under/over-growing of burnt area. Validation data covered a range of burnt areas from ten independent Path/Rows with ten images sampled across a range of burnt area fractions per Path/Row (n = 100). The results for burnt area mapping demonstrated an average producer's accuracy of 85% (range of 28 to 100% for individual images) and average user's accuracy of 71% (range of 4 to 99% for individual images). A morphological dilation of one pixel restricted to locations exhibiting a decline in B45 over time, increased the producer's accuracy by 4% but reduced the user's accuracy by 8%. The total accuracy for the burnt area classification was greater than 99%, however this is more a reflection of the small fraction of landscape represented by burnt area rather than the ability to detect burnt area. Areas frequently misclassified were related to areas of high spectral/land use change which included areas of cropping, frequently inundated land, and moisture/ground cover variations over dark soils. In this study, we applied a crop and water mask to minimise commission errors. Significantly, the results of this study demonstrate that an automated time series method for mapping burnt areas can be successfully applied across a diversity of land cover types. The method may be applied in similar savanna dominated environments but is likely to require modification to be applicable in other landscapes.
Article
Full-text available
Prescribed fire is increasingly used to inhibit woody encroachment into fire-dependent ecosystems, yet its effects on other processes influencing invasion are poorly understood. We investigated how fire influences exotic woody invasion through its effects on granivore activity, and whether these effects depend on the habitat in which seed predation takes place. We quantified seed removal for four species of exotic woody plants (Albizia julibrissin, Elaeagnus umbellata, Melia azedarach and Triadica sebifera) in 17 sites in longleaf pine savanna that varied in time since fire (one or three growing seasons post-fire) in the sandhills region of North Carolina, USA. Within each site, we established paired plots in upland and upland-wetland ecotone communities and presented seeds in depots that allowed either arthropod, or arthropod and small vertebrate access. We found that differences in seed removal with time since fire were contingent on habitat and granivore community. In ecotones, three of four species had higher proportions of seeds removed from plots that were three growing seasons post-fire than plots one growing season post-fire, whereas only T. sebifera showed this pattern in upland habitat. Allowing vertebrate granivores access to seeds enhanced seed removal, and this effect was strongest in ecotone habitat. While granivores removed seeds of all four plant species, removal of E. umbellata was significantly higher than that of the other species, suggesting that granivores exhibited seed selection. These findings suggest that ecotone habitats in this system experience greater seed removal than upland habitats, particularly as time since fire increases, and differences are mainly due to the activity of vertebrate granivores. Such differences in seed removal, together with seed selection, may contribute to variation in exotic woody invasion of longleaf pine savannas.
Article
Full-text available
Ecological systems often exhibit resilient states that are maintained through negative feedbacks. In ponderosa pine forests, fire historically represented the negative feedback mechanism that maintained ecosystem resilience; fire exclusion reduced that resilience, predisposing the transition to an alternative ecosystem state upon reintroduction of fire. We evaluated the effects of reintroduced frequent wildfire in unlogged, fire-excluded, ponderosa pine forest in the Bob Marshall Wilderness, Montana, USA. Initial reintroduction of fire in 2003 reduced tree density and consumed surface fuels, but also stimulated establishment of a dense cohort of lodgepole pine, maintaining a trajectory toward an alternative state. Resumption of a frequent fire regime by a second fire in 2011 restored a low-density forest dominated by large-diameter ponderosa pine by eliminating many regenerating lodgepole pines and by continuing to remove surface fuels and small-diameter lodgepole pine and Douglas-fir that established during the fire suppression era. Our data demonstrate that some unlogged, fire-excluded, ponderosa pine forests possess latent resilience to reintroduced fire. A passive model of simply allowing lightning-ignited fires to burn appears to be a viable approach to restoration of such forests.
Article
Full-text available
We examined factors likely to be important in management of rare flora of the longleaf pine-wiregrass ecosystem on Fort Bragg and Camp Mackall Military Reservations in North Carolina, USA. We documented 1,268 occurrences of thirty-six rare plant species during two inventories conducted from 1991 to 1993 and 1998 to 1999. Of these occurrences, 891 (70.3%) persisted, 258 (20.3%) went extinct, and 119 (9.4%) colonized between the first and second inventories; extinctions exceeded colonizations by 139 occurrences. We used analyses of contingency table frequencies and logistic regressions to test hypotheses about temporal responses of local occurrences of rare flora. We found statistically significant effects of fire frequency on the colonization and extinction of rare flora; extinctions declined and persistences increased with increasing area occupied and stem number. Our analyses emphasize the importance of fire and population size for the conservation of rare flora in this landscape.
Article
Full-text available
We report onthe numerical separation ofburned and unburned vegeta- tion classes using diVerent bi-spectral spaces, based on the analysis of spectro- radiometric data collected in situ and convolved to é ve spectral bands at red to mid-infrared (MIR) wavelengths. A combination of two MIR bands was found to have strong spectral separation of burned and unburned samples. Using these bands, a spectral index was formulated which is highly sensitive to spectral changes due to burning and relatively insensitive to intrinsic variability. Results have implications for the remote sensing of burned shrub-savannah using bands available on high- and low-spatial resolution sensors, in particular, Landsat TM and MODIS.
Article
Full-text available
In this work, I have used life-history and demographic data to define fire return intervals for several types of Florida scrub, a xeric shrubland where fire is the dominant ecological disturbance but where fire suppression is a major issue. The datasets combine chronosequence and longitudinal approaches at community and population levels. Resprouting shrubs, which dominate most types of Florida scrub, recover rapidly after fires (although their limits under frequent fires are not well known) and also increasingly dominate long-unburned areas. These dominant shrubs can prosper over a range of fire return intervals. Obligate-seeding scrub plants, which often have persistent seed banks, can be eliminated by frequent fire but often decline with infrequent fire. Population viability analyses of habitat specialists offer more precision in suggesting ranges of appropriate fire return intervals. For two types of Florida scrub (rosemary scrub and oak-hickory scrub), plant-population viability analyses narrow the interval and suggest more frequent fires than do previous recommendations, at intervals of 15-30 and 5-12 years, respectively. Variation in fire regimes in time and space (pyrodiversity) is recommended as a bet-hedging fire-management strategy and to allow co-existence of species with disparate life histories.
Article
Full-text available
Remote sensing techniques are specially suitable to detect and to map areas affected by forest fires. In this work, Landsat 5 Thematic Mapper (TM) data has been used to study a number of forest fires that occurred in the province of Valencia (Spain) and to monitor the vegetation regeneration over burnt areas.A reference area (non‐burnt forest) was established to assess the change produced by fire. The radiance in the thermal band (10.4–12.5 μm) and the normalized difference in reflectance between near IR (0.76–0.90 μm) and middle IR (2.08–2.35 μm) were the most suitable parameters to map burnt areas. This index can also be used for monitoring vegetation regeneration in burnt areas. About a month after the fire, the burns show temperatures of 5–6 °C higher than those found in the reference area, and the vegetation index shows negative values whereas the reference area values remain positive. The differences between the burns and the reference area for the vegetation index decrease with time as vegetation regenerates.
Article
Full-text available
This study is a compilation of vascular flora data from previous reports and current findings from 1965 through 2003 of Fort Bragg Military Reservation and Weymouth Woods Sandhills Nature Preserve, located in the Sandhills region of North Carolina. Vascular plants are divided into four major groups: Pteridophytes, Gymnosperms, Angiosperms (monocots) and Angiosperms (dicots). Reported for the study area are 143 families, 490 genera, and 1,206 species and infraspecific taxa, of which 203 (16.8%) are alien or adventive. Sixty one species are rare, of which three are federally endangered: Lysimachia asperulifolia, Rhus michauxii, and Schwalbea americana. The following five species are currently known in North Carolina only from study area populations: Carex tenax, Hypericum fasciculatum, Pteroglossaspis ecristata, Rhynchospora crinipes, and Warea cuneifolia. This study documents the importance of Fort Bragg Military Reservation and Weymouth Woods Preserve in contributing to the floristic diversity of the longleaf pine-wiregrass ecosystem.
Article
Full-text available
Lightning- and human-ignited fires have shaped many of Florida's natural communities. Plant persistence in these habitats requires adaptation to survive high temperatures, ability to respond to aboveground tissue loss, or avoidance of these conditions through survival in unburned patches or in the soil seed bank. Limited distribution of many rare species in increasingly fragmented habitats and uncertainty of whether rare species tolerate, require, or persist through avoidance of fire hampers fire management when it is possible and threatens long-term viability of plants. We review up to 14 years of monitoring data for 18 rare plant species from 14 families occurring on sandhill, scrub, pine rockland, and mixed deciduous hardwood communities across Florida. For a number of variables measured, including density, frequency, flowering, and recruitment, 50% of the species showed significant positive responses to burning and 50% of the species showed a neutral or no response. None of the species were unable to recover post-burn. These results suggest that prescribed fire in these pyrogenic habitats need not be delayed until species-specific responses to fire are understood.
Article
Full-text available
We studied the effects of experimental fire regimes, (dormant season fire, growing season fire, growing season mowing and control, i.e., no experimental treatment) on populations of the USA federally endangered, Schwalbea americana L. between 1992 and 1996. Although this species occurs in fire-maintained habitat in the Southeastern USA, there is concern about the use of fire for such rare populations. The purpose of the study was to examine how seasonal timing of fire and fire suppression affect population demography, flowering phenology and spatial distribution; to identify modes of persistence associated with fire regimes; and to determine if summer mowing provides a management alternative to fire. Fire-induced flowering was demonstrated in this species. Seasonal timing of burns appears to have relatively little consequence on population structure or spatial extent, but alters flowering phenology. Burning, regardless of season, resulted in increased population density and expansion in areal extent. Two possible mechanisms of persistence between fire events were identified including regression from reproductive stage to vegetative stage in the absence of fire and dormancy of individual plants for one or more seasons. Growing season mowing does not appear to be an adequate substitute for burning.
Article
Full-text available
Theoretically, fire–tree cover feedbacks can maintain savanna and forest as alternative stable states. However, the global extent of fire-driven discontinuities in tree cover is unknown, especially accounting for seasonality and soils. We use tree cover, climate, fire, and soils data sets to show that tree cover is globally discontinuous. Climate influences tree cover globally but, at intermediate rainfall (1000 to 2500 millimeters) with mild seasonality (less than 7 months), tree cover is bimodal, and only fire differentiates between savanna and forest. These may be alternative states over large areas, including parts of Amazonia and the Congo. Changes in biome distributions, whether at the cost of savanna (due to fragmentation) or forest (due to climate), will be neither smooth nor easily reversible.
Article
Full-text available
Empirical models relating forest attributes to remotely sensed metrics are widespread in the literature and underpin many of our efforts to map forest structure across complex landscapes. In this study we compared empirical models relating Landsat reflectance to forest age across Oregon using two alternate sets of ground data: one from a large (n approximately 1500) systematic forest inventory and another from a smaller set of plots (n < 50) deliberately selected to represent pure conditions along predefined structural gradients. Models built with the smaller set of targeted ground data resulted in lower plot-level mapping error (root mean square error) and higher apparent explanatory power (R2) than those built with the larger, more widely distributed inventory data. However, in two of the three ecoregions considered, predictions derived from models built with the smaller ground data set displayed a bias relative to those built with the larger but noisier inventory data. A modeling exercise, wherein mapped forest age was translated into carbon, demonstrated how nonlinear ecological models can magnify these prediction biases over landscapes. From this study, it is clear that for mapping purposes, inventory data are superior to project-specific data sets if those data sets are not representative of the full region over which mapping is to be done.
Article
Biodiversity is being lost rapidly due to anthropogenic changes in land use, climate, and other environmental conditions. In fire‐maintained ecosystems, altered fire regimes accelerate native species loss—community disassembly—and promote recruitment of fire‐sensitive species. In this study, we ask whether fire suppression results in changes over time in functional trait composition of ground‐layer species and whether these changes differ in longleaf pine savannas invaded by hardwoods from those invaded by sand pines. Five Floridian locations on the Southeastern Coastal Plain of the United States. At each location we selected a fire‐maintained and fire‐suppressed savanna and measured percent plant cover by ground‐story species in 1,000 m2 plots. For 102 of these species, we measured 10 functional traits—height, growth form, specific leaf area, leaf dry matter content (LDMC), leaf water content, leaf ignition time, leaf mass consumed by fire, light compensation points (LCPs), nonstructural carbohydrate concentrations in under‐ground organs, and seed mass. Fire exclusion was associated with reductions in both functional diversity and species richness. We identified 38 species exclusive to frequently burned sites, these species showed high LDMC values, high LCPs, high leaf mass consumed, and low leaf ignition time values. Lack of fire was associated with loss of 6 of the 12 C4 native grass species. Species and functional trait composition were affected by both time‐since‐fire and whether post‐fire communities were invaded by broadleaved trees (Quercus spp. and Liquidambar styraciflua) or by sand pine (Pinus clausa). We demonstrated the effects of altered disturbance regimes on savanna plant species and functional trait composition. This trait‐based approach advanced our understanding of how altered disturbance regimes can alter plant communities. Although 38 shade intolerant and flammable native savanna species were absent by 10 years since fire, 12 such species persisted even 40 years after fire exclusion.
Article
Fire-promoting, open-canopy ecosystems are under threat of conversion to a fire-deterring, closed-canopy condition due to woody encroachment. This conversion of vegetation structure has been fostered by introduced woody plant species. We performed a field experiment to quantify growth, survival, and establishment success of six invasive, woody species along a managed longleaf pine savanna–wetland gradient in the Sandhills of North Carolina, USA. We investigated the effects of prescribed fire, fire history, dispersal, and abiotic conditions on the invasibility of sites along the gradient. Across 18 study sites, seeds of the six woody species were sown using three sowing methods that mimicked primary and secondary dispersal; each site contained paired plots located in savanna and savanna-wetland ecotone vegetation communities. We identified sowing treatment, abiotic conditions, seedling size, and prescribed fire as important factors for controlling woody invasion, as they prevented 5 of 6 study species from establishing in the landscape. However, the landscape was not immune to invasion. At the end of the 42-month study period, three species had established in unburned sites. In sites burned after seedling emergence, only one species, Pyrus calleryana, survived and established. We found P. calleryana survival and establishment to be a function of seedling size, soil humic matter content, and sowing treatment. Successful invasion and establishment of woody individuals in open-canopied systems increases the likelihood of fire-deterrence and further woody encroachment, threatening ecosystem integrity.
Article
The Carolina Sandhills is a physiographic region of the Atlantic Coastal Plain province in the southeastern United States. In Chesterfield County (South Carolina), the surficial sand of this region is the Pinehurst Formation, which is interpreted as eolian sand derived from the underlying Cretaceous Middendorf Formation. This sand has yielded three clusters of optically stimulated luminescence ages: (1) 75 to 37 thousand years ago (ka), coincident with growth of the Laurentide Ice Sheet; (2) 28 to 18 ka, coincident with the last glacial maximum (LGM); and (3) 12 to 6 ka, mostly coincident with the Younger Dryas through final collapse of the Laurentide Ice Sheet. Relict dune morphologies are consistent with winds from the west or northwest, coincident with modern and inferred LGM January wind directions. Sand sheets are more common than dunes because of effects of coarse grain size (mean range: 0.35–0.59 mm) and vegetation. The coarse grain size would have required LGM wind velocities of at least 4–6 m/sec, accounting for effects of colder air temperatures on eolian sand transport. The eolian interpretation of the Carolina Sandhills is consistent with other evidence for eolian activity in the southeastern United States during the last glaciation.
Article
Fire has a major impact on the structure and function of many ecosystems globally. Pyrodiversity, the diversity of fires within a region (where diversity is based on fire characteristics such as extent, severity, and frequency), has been hypothesized to promote biodiversity, but changing climate and land management practices have eroded pyrodiversity. To assess whether changes in pyrodiversity will have impacts on ecological communities, we must first understand the mechanisms that might enable pyrodiversity to sustain biodiversity, and how such changes might interact with other disturbances such as drought. Focusing on plant-pollinator communities in mixed-conifer forest with frequent fire in Yosemite National Park, California, we examine how pyrodiversity, combined with drought intensity, influences those communities. We find that pyrodiversity is positively related to the richness of the pollinators, flowering plants, and plant-pollinator interactions. On average, a 5% increase in pyrodiversity led to the gain of approximately one pollinator and one flowering plant species and nearly two interactions. We also find that a diversity of fire characteristics contributes to the spatial heterogeneity (β-diversity) of plant and pollinator communities. Lastly, we find evidence that fire diversity buffers pollinator communities against the effects of drought-induced floral resource scarcity. Fire diversity is thus important for the maintenance of flowering plant and pollinator diversity and predicted shifts in fire regimes to include less pyrodiversity compounded with increasing drought occurrence will negatively influence the richness of these communities in this and other forested ecosystems. In addition, lower heterogeneity of fire severity may act to reduce spatial turnover of plant-pollinator communities. The heterogeneity of community composition is a primary determinant of the total species diversity present in a landscape, and thus, lower pyrodiversity may negatively affect the richness of plant-pollinator communities across large spatial scales.
Article
Random forests are a combination of tree predictors such that each tree depends on the values of a random vector sampled independently and with the same distribution for all trees in the forest. The generalization error for forests converges a.s. to a limit as the number of trees in the forest becomes large. The generalization error of a forest of tree classifiers depends on the strength of the individual trees in the forest and the correlation between them. Using a random selection of features to split each node yields error rates that compare favorably to Adaboost (Y. Freund & R. Schapire, Machine Learning: Proceedings of the Thirteenth International conference, ∗∗∗, 148–156), but are more robust with respect to noise. Internal estimates monitor error, strength, and correlation and these are used to show the response to increasing the number of features used in the splitting. Internal estimates are also used to measure variable importance. These ideas are also applicable to regression.
Article
A flatwoods longleaf pine wiregrass ecosystem, which regenerated naturally following wildfire in 1942, on the Coastal Plain of southern Georgia was treated over a period of four decades with prescribed fire at annual, biennial and triennial intervals during the winter dormant season. Burning caused substantial changes in the understory plant community, with significant reductions in the foliar cover of Ilex glabra in the shrub layer resulting in corresponding increases in the cover of Vaccinium myrsinites, Sporobolus curtissii, Aristida stricta and Andropogon spp. Understory plant species richness, diversity and evenness also increased as a result of periodic fire. Dormant-season burning decreased the cover of litter on the forest floor and significantly increased the standing biomass of A. stricta, S. curtissii, Andropogon spp., all other grasses and all forbs. Recurrent fire also prevented the development of a vigorous midstory, that impedes understory growth and poses a serious fire hazard to the stand. Overstory trees were largely unaffected by burning. Historical light grazing on the site produced no measurable effects on the plant community. Findings suggest that the biennial burning interval results in declines of I. glabra in the shrub layer and litter cover on the forest floor, leading to the largest increases in understory plant species richness and diversity and the biomass productivity of grasses and forbs. Although flatwoods plant communities evolved in environments characterized by growing-season fires of variable frequency, long-term application of dormant-season fire is also recommended as a useful option for sustaining resource values in this and similar longleaf pine wiregrass ecosystems.
Article
Fire can have dramatic effects on the vital rates of plant species and has been used successfully for management in a number of ecosystems. However, the demographic response of species to fire in fire-dependent ecosystems is variable, making it important to study the effects of fire on rare and threatened species. We quantified the effects of fire on Astragalus michauxii and Pyxidanthera brevifolia, two rare endemics of the longleaf pine-wiregrass ecosystem of the southeastern USA, by means of periodic matrix models to project the effect of fire frequency on population growth. In contrast to many species in the longleaf pine-wiregrass ecosystem, fire had short-term negative effects on both species, causing reductions in survival, size, flowering, and fruit production. Relative to the three-year fire intervals to which the study populations are currently exposed, more frequent burning is projected to cause population decline, with the most dramatic effects under annual burning. Although the current longleaf pine-wiregrass ecosystem is fire dependent and has experienced frequent fire for at least several thousand years, we propose that the two endemic species may be remnants from a past vegetation assemblage that experienced less frequent fire and thus may be adapted to longer fire-return intervals compared to other species currently in the ecosystem. Despite the short-term negative effects of fire on the vital rates of these species, longer-term benefits such as reduction of woody encroachment and litter removal may be important for the ultimate success of the species.
Article
Understanding how a species persists under a particular fire regime requires knowledge of the response to fire of individual plants. However, categorising the fire response of a species solely based on known responses of individual plants can be misleading when predicting a population response. In the present study, we sought to determine the fire responses of several Leucopogon species at the population level, including the threatened L. exolasius.We found that, whilst all species studied were obligate seeders, the population responses of species to firewere dependent upon fire intensity and patchiness. Results showed first that lowintensity fireswere significantly patchier than higher intensity fires. Second, the proportion of plants killed within a population decreased with increased fire patchiness.We also assessed how populations were structured and found that stands were multi-aged at most sites, and did not have a single-aged structure, which is often assumed for obligate seeders. Both spatial complexity within the fire regime leading to adult plant persistence, and inter-fire recruitment, contributed to the multi-aged structure. It is possible that these Leucopogon species are gap recruiters, and may tolerate fire rather than be specifically adapted to it. Inter-fire recruitment may enable L. exolasius populations to persist for a much longer fire-free period than many other species in the region.
Article
Cited By (since 1996): 1, Export Date: 17 January 2013, Source: Scopus, CODEN: RSEEA, doi: 10.1016/j.rse.2011.03.010, Language of Original Document: English, Correspondence Address: Disney, M.I.; Department of Geography, University College London, Gower Street, London, WC1E 6BT, United Kingdom; email: mdisney@geog.ucl.ac.uk, References: Archibald, S., Scholes, R.J., Roy, D., Roberts, G., Boschetti, L., Southern African fire regimes as revealed by remote sensing (2010) International Journal of Wildland Fire, 19 (7), pp. 861-878;
Article
Atmospherically corrected Moderate Resolution Imaging Spectroradiometer (MODIS) data have been used to measure the changes in surface reflectance induced by fires. To account for observation geometry effects a kernel driven bi-directional reflectance factor model was applied. Whereas the blue, green, red and shortwave infrared bands show no consistent behaviour, the near-infrared bands almost always show a strong reduction in reflectance. An angular dependence of the change in reflectance was not found in this study. Different bio-geographical regions exhibit different spectral reflectance changes due to the different types of fuel being burnt (green/living versus dry/dead vegetation). This difference is also reflected in the seasonality of the green, red, near-infrared and shortwave infrared bands for the tropics. The conclusion of this study is that the near-infrared bands are the most suitable bands for an automatic burnt area mapping algorithm using optical, reflective remote sensing data. The results also suggest that satellite remote sensing might be able provide additional information about burning conditions which are strongly affecting greenhouse gas emissions.
Article
Questions: Has fire suppression relaxed barriers to the exchange of species between savanna and forest? Do all species or a subset of species participate in this exchange? Would current vegetation structure persist if fire suppression were to cease? Location: A gallery forest edge in the Cerrado region of central Brazil that burned only once in the past 35 years. Methods: Density of tree seedlings, saplings and adults, leaf area index (LAI), tree basal area and diameter were surveyed in 12, 10 m × 70 m transects centred on and perpendicular to the forest–savanna boundary. Community composition was assessed using non-metric multi-dimensional scaling (NMDS). Results: Basal area and LAI declined substantially from forest to savanna, with an associated shift in species composition. Savanna tree species were nearly absent in the forest, but accounted for the majority of stems in the savanna. In contrast, forest species comprised 14% of adults and more than one-third of juveniles in the savanna. Despite the high diversity of trees (85 species) in the forest, five species play a particularly large role in this initial phase of forest expansion. Reintroduction of fire, however, would result in widespread topkill of juveniles and the majority of adult forest trees, thereby interrupting the succession towards forest. Conclusions: After 35 years during which the site burned only once, the savanna still remains dominated by savanna species. Nevertheless, the dominance of forest juveniles in border and savanna tree communities suggests that with a continued policy of fire suppression, the forest will continue to expand.
Article
Question: How frequent and variable were fire disturbances in longleaf pine ecosystems? Has the frequency and seasonality of fire events changed during the past few centuries? Location: Kisatchie National Forest, Western Gulf Coastal Plain, longleaf pine–bluestem ecosystem, in relatively rough topography adjacent to the Red River, Louisiana, USA. Methods: Cross-sections of 19 remnant pines exhibiting 190 fire scars were collected from a 1.2-km2 area. Tree-rings and fire scars were precisely dated and analysed for the purpose of characterizing past changes in fire and tree growth. Temporal variability in fire occurrence and seasonality was described for the pre- and post-European settlement periods. Seasonality of historic fires was determined by the scar position within the rings. The relationship between fire and drought was investigated using correlation and superposed epoch analysis. Results: The mean fire return interval for the period 1650-1905 was 2.2 years (range 0.5 to 12 yr). Significant new findings include: evidence for years of biannual burning, temporal variability in fire seasonality, an increase in fire frequency and percentage of trees scarred circa 1790, and synchronous growth suppression and subsequent release of trees coinciding with land-use changes near the turn of the 20th century. Drought conditions appeared unrelated to the occurrence of fire events or fire seasonality. Conclusions: Multi-century fire history records from longleaf pine ecosystems are difficult to obtain due to historic land-use practices and the species high resistance to scarring; however, our results indicate potential for reconstructing detailed fire histories in this ecosystem. Fire scars quantitatively documented one of the most frequent fire regimes known. Fire regime information, such as the temporal variability in fire intervals, prevalence of late-growing season fire events and biannual burning, provide a new perspective on the dynamics of longleaf pine fire regimes.
Article
We examined the effects of fire on population growth rate and extinction probability of a rare prairie plant. Lomatium bradshawii (Apiaceae) is an endangered species of western Oregon and Washington prairies that were frequently burned by Native Amer-icans prior to the late 1800s. Using data from mapped plants in two populations subjected to different fall burning frequencies over a six-year period, we constructed stochastic tran-sition matrix models to evaluate the effects of fire on the viability of the populations. Stochasticity was incorporated into the models through two methods performed by widely used computer programs: randomly shuffling whole annual matrices (using POPPROJ2), or selecting each matrix element from a distribution with observed mean and variance (with RAMAS/stage). Log-linear analysis was used to test for site and treatment effects on transition rates. Unburned plots had stochastic population growth rates (s) of 0.905–0.927, depending on the site and stochastic method. Burning twice in six years increased s to 0.946–1.091, and three burns yielded s values of 0.996–1.173. The risk of extinction (100 yr, extinction defined as 10 individuals) was very high (97–99%) in the absence of fire for both methods and sites, but with two burns it declined to 57% and 73% for matrix selection and element selection, respectively, at one site, and 1% (both methods) at the other. Extinction probability was very low (1%) for both methods and sites after three burns. Although different methods of incorporating stochasticity gave slightly different quantitative results, they were qualitatively similar. Both site and treatment had significant effects on transition rates in the log-linear analyses. Elasticity analysis suggested that burning decreased the sensitivity of population growth rate to mid-sized plants and increased the importance of seedlings. Fire was an effective tool for maintaining viable populations of this species, although the strength of its effects differed between locations.
Article
The longleaf pine-grassland (Pinus palustris-Poaceae) ecosystem occupied over 30 million ha in the southeastern United States at the time of European discovery. Frequent low- to moderate-intensity surface fires ignited by both lightning and native Americans sustained open diverse stands in a fire climax and prevented succession to mixed hardwood forests. Disruption of pre-historical and historical fire regimes, coupled with land conversion, urbanization, and other factors, is responsible for the rapid decline of the ecosystem. Today only about 1.2 million ha remain, much in isolated fragments. Primarily because of habitat loss, many animal and plant species associated with longleaf forests are now rare or in decline. Restoration ecologists and managers face a daunting challenge—recreating an ecosystem, in the face of chronic cumulative stress from human activities, that varied widely over temporal and spatial scales. Key restoration factors include: (1) development of a general understanding of the historical condition of the longleaf ecosystem, especially unusual or unique communities and habitats embedded in the general fabric of the larger ecosystem, (2) initiation and expansion of a fire regime, where feasible, similar to that which historically shaped the ecosystem, (3) maintenance/enhancement of herbaceous diversity, (4) continued research on habitat requirements and distribution of rare species, and (5) encouragement of a multi-owner partnership approach to promote conservation across the landscape. Landowners and the public must be educated about the values of the longleaf pine-grassland ecosystem and develop a conservation ethic that considers aesthetics, wildlife, and biodiversity, in addition to economics, if the ecosystem is to be restored. Most forestry practices used to manage and restore longleaf forests are of low short-term risk to rare species in this ecosystem. The benefits of active management usually far outweigh the long-term risks associated with no management.