Mountain pine beetle, Dendroctonus ponderosae Hopkins, is among the most significant forest insect pests, particularly in lodgepole pine, Pinus contorta Dougl. ex Loud., forests in western North America. We established a network of 125, 0.081-ha circular plots in P. contorta forests in Colorado, Idaho, Montana, Utah, and Wyoming to document tree mortality and stand-level responses to D. ponderosae outbreaks. We found notable reductions in several key metrics but also observed significant variability in per-year (0–71%) and cumulative mortality (2–87%). We used this long-term monitoring dataset and employed generalized linear mixed effects modeling to determine the biophysical drivers of the variation in P. contorta mortality. Models were fitted to the entire (2005–2024) dataset, and to two truncated datasets: outbreak years only (2005–2012) and maximum per year P. contorta mortality per plot. Explanatory variables included tree, plot, and environmental factors (temperature and water availability). Our models highlighted several significant drivers of P. contorta mortality. The most significant and consistent were individual tree dbh and quadratic mean diameter (QMD). Larger individual tree dbh and larger QMD positively influenced the probability of P. contorta mortality. Tree dbh interacted significantly with water availability variables, particularly November-March snowpack. Increased snowpack and greater densities of nonhost tree species decreased mortality. These results suggest that management efforts should account for the frequency of large-diameter P. contorta and that density reduction and/or increasing tree species (nonhost) and structural diversity may provide mitigation strategies for reducing P. contorta mortality attributed to D. ponderosae in the Intermountain West.

Accurate mapping of post-fire surviving trees is important for tracking forest recovery and prioritizing land management decisions. Satellite-based remote sensing is an effective method to assess post-fire forest conditions. Traditionally, differenced satellite-derived burn severity indices are computed by differencing one year pre- and post-fire spectral reflectance values. Differenced burn severity indices are useful for quantifying and mapping the magnitude of ecological change, but their application to detecting and mapping post-fire live trees may not be as appropriate, particularly for delayed tree mortality. Delayed tree mortality ("delayed mortality") is a phenomenon where trees that initially survive fire then die over an extended period (between one and five years), and it can be challenging to measure and predict. In this study, we demonstrate the potential of mapping delayed mortality using readily available remotely sensed imagery alone. We used random forest models to detect postfire live trees using 10-m resolution Sentinel-2 data at one-, three-, and five-years post-fire for four fires in the southern Sierra Nevada, California, USA. Using imagery from the National Agriculture Imagery Program (NAIP; 60-cm resolution), we manually classified live tree presence in 6000 Sentinel-2 pixels (500 pixels for each fireyear combination) to calibrate and validate models. Sentinel-2 based model accuracies ranged from 65 % to 86 % with F-scores ranging from 0.52 to 0.86, and their predictions of live pixel area were on average 44 % lower than inferred from more traditional indices such as relative differenced normalized burn ratio (RdNBR). This work represents a promising first step in using freely available post-fire spectral reflectance imagery to detect live trees over an extended period to support post-fire management.

Abstract Wildfires in California’s Sierra Nevada during 2020–2021 killed giant sequoias ( Sequoiadendron giganteum ) at rates unseen for millennia, underscoring the vulnerability of highly fire-adapted trees to ongoing environmental change. Following a century of fire exclusion and fuel accumulation, the effectiveness of prescribed burns in reducing giant sequoia mortality from wildfire remained poorly quantified. Here we estimate mortality outcomes for 26,403 giant sequoias across 19 groves in Sequoia and Kings Canyon national parks following the Castle (2020) and KNP Complex (2021) wildfires using a Bayesian framework. We map tree mortality using a deep learning classifier integrating 3 m PlanetScope imagery, airborne lidar, and field observations. From an estimated 7,974 sequoia deaths (95% Bayesian credible interval (CI): 7,555–8,430), corresponding to 30.2% mortality (CI: 28.6–31.9%), we find previous prescribed burns (≤10 years prior) reduced mortality odds by 77% (CI: 69–83%), making treated trees nearly four times more likely to survive. Counterfactual simulations suggest that prescribed burns prevented at least 1,888 (CI: 1,487–2,302) deaths, and universal treatment would have saved an additional 3,888 (CI: 3,236–4,580) giant sequoias. These results show that prescribed burns substantially improve survival during extreme wildfires, offering guidance for conserving long-lived, fire-adapted forests under intensifying fire regimes.

High-severity fire activity in western US dry forests has increased in recent decades, with models projecting extensive declines in tree recruitment and tree species diversity for mixed-conifer forests. While burn severity, seed limitation, and precipitation are known filters of post-fire recovery, seed predation by small mammals may further influence regeneration by reducing conifer seed availability. We quantified how seed removal varies with fire severity, salvage logging, and distance into high-severity patches, as well as across conifer species common to mixed-conifer forests. We measured seed removal rates for four conifer species 2 years following a large wildfire. Sugar pine (Pinus lambertiana), ponderosa pine (Pinus ponderosa), and Douglas-fir (Pseudotsuga menziesii) seeds experienced significantly higher removal rates in high-severity areas relative to low-severity areas, while white fir (Abies concolor) removal remained low across all conditions. Removal of non-fir species in salvage-logged sites was intermediate but not significantly different from other overstory conditions, and did not vary with distance from surviving trees. Mice (Peromyscus sp.) were the primary seed predators in high-severity sites. These findings are based on a single season of data with a modest sample size. Findings indicate that fire severity amplifies underlying foraging preferences against white fir. The positive effects of fire severity on seed removal may help explain prior observations of low overall conifer recruitment and disproportionate white fir dominance following high-severity fire in mixed-conifer forests.

Fuel treatments are crucial for reducing wildfire hazard, especially as severe wildfires increase across western United States (US) dry conifer forests. While many studies have documented the effectiveness of fuel treatments in reducing future wildfire severity, few have synthesized data to predict post-treatment fuel loads for major classes of fuels. We conducted a Bayesian meta-analysis using 1932 observations from 65 published papers in western US dry conifer forests to (1) evaluate the short-term effects of different fuel treatments on fuel loading and overstory structure and (2) characterize patterns of post-treatment fuel loading and overstory structure across multiple fuel components. Treatments included thinning followed by prescribed burning (THIN+BURN), thinning only (THIN), and first-entry prescribed fire only (BURN). Our results show that treatments effectively reduce fuel loads, but outcomes vary based on treatment type, forest type, and initial stand conditions. THIN+BURN treatments were most effective in reducing overstory fuel loads while preventing surface fuel buildup, whereas BURN treatments were the most effective at reducing surface fuel loads, even after a single entry. Our findings underscore the importance of treatment type and pre-treatment stand conditions in influencing fuel reduction outcomes. Fuel treatments, especially in landscapes with heavy fuel loads, offer a valuable tool for moderating wildfire severity, reducing fire risk, and promoting forest restoration. Our synthesis of posttreatment fuel loads provides important insights for assessing forest vulnerability, improving fire behavior model estimates, and informing wildfire management strategies in a changing climate.

Across much of the semiarid conifer forests of western North America ("dry conifer forests"), the dominant tree species are non-serotinous, lack soil seedbanks, and rarely disperse seeds much farther than 100 m, so tree regeneration in large, high-severity burned patches is expected to be highly seed-limited. Conifer seedlings do, however, sometimes establish at high densities deep within high-severity patches in these forests, implying that seeds can sometimes survive intense wildfire even when all overstory trees die. Does seed survival in the canopies of non-serotinous trees provide an unexpected source of forest resilience? To answer this question, we surveyed tree survival, fire severity, and seedling abundance across two very large wildfires in the first year after fire. Several of the study species had a good seed cone production year at the time of the fires. We stratified many of our plots deep within high-severity patches far from surviving trees, where existing models predict regeneration failure due to lack of viable seeds. Contrary to such expectations, we found that conifer seedling densities in these areas were generally far greater than needed to replace the fire-killed stand and sometimes approached seedling densities observed near surviving trees. Seedling densities in high-severity areas far from surviving trees correlated negatively with local burn intensity (canopy foliage consumption), supporting the idea that the seeds originated locally and highlighting a critical driver of post-fire recovery that is easily missed by traditional surveys conducted >2 years following fire. Seedling density was also strongly associated with burn date, suggesting that persistence of viable canopy seeds depends on synchrony between wildfire and cone ripening dates. Together, our results demonstrate that under the right conditions, canopy seed survival can lead to dense seedling establishment across large severely burned areas and may substantially support the resilience of dry conifer forests to the uncharacteristically severe fires that are becoming increasingly prevalent in this system.

Identifying live tree presence following wildfire is important for burn damage assessments and decision making, as these trees serve as seed sources for recovery. Satellite-based remote sensing offers an efficient means to assess burn severity with products representing vegetation greenness and char/ash presence and their change from pre- to post-fire imagery. While effective at assessing burn severity (e.g. ecosystem change), there remain limitations in identifying fire refugia (surviving trees), due to the difficulty of teasing apart different green vegetation types (e.g. trees, shrubs, grasses). In this paper, we use 10 m Sentinel-2 satellite data to predict live tree presence across three sites impacted by the 2021 California fire season. We used vegetation indices (VIs) from post-fire imagery (normalized difference vegetation index [NDVI], normalized burn ratio [NBR], normalized difference water index [NDWI], visible atmospherically resistant index [VARI], and burn area index [BAI]), differential VIs from pre- and post-fire imagery (dNDVI, dNBR, RdNBR, dNDWI, dVARI), and direct reflectance bands (all bands model; visible, near-infrared, and shortwave infrared; B1–B12) to predict live tree presence via random forest modeling. To calibrate and validate the random forest models, we photointerpreted ∼2300 pixels per fire region using 2022 National Agriculture Imagery Program imagery. We performed additional field-based validation using tree presence/absence data two years post-fire ( n = 296 observations across two sites). At the site level, the all bands model outperformed the vegetation index-based models (80%–85% vs 65%–79% accuracy). Errors were mainly false positives attributed to pixels with green understory vegetation but no live trees. In cross-site inference, which involved pooling two sites for model calibration to test on the third site, the all bands model retained good performance (76%–81% accuracy). Evaluation against field survey data demonstrated a larger range of performance (50%–87% accuracy) that highlights limitations based on tree isolation and crown percent greenness. Relative to differential-based VIs, our results highlight potential advantages of using post-fire Sentinel-2 imagery and random forest modeling for identifying live tree presence and scaling to full fire extents.

With dramatic increases in both area burned and fire severity in Western North American conifer forests, the demand for postfire replanting greatly exceeds land managers' capacity. Despite the importance of tree planting for forest recovery in many areas, it remains unclear how environmental variation and planting timing affect tree planting success relative to passive natural tree regeneration, or how to optimize limited planting resources by focusing on the right places at the right time. To address this gap, we surveyed replanting success across five fires in California Sierra Nevada mixed conifer forest that had been partially replanted with conifer species after intense wildfire. We selected these fires to contain substantial variation in environmental conditions (temperature, elevation, other topographic contrasts) and planting timing, while being as consistent as possible in postfire management. At each fire, we surveyed randomly located 400 m2 circular plots in planted and nearby unplanted areas (total 182 plots), counting seedlings by species as well as shrub cover and other environmental variables. Using mixed models to analyze the data, we found that passive natural regeneration is weaker (<50 seedlings/ha) in hotter, drier sites and that active tree planting can provide a boost (up to 200 %) to forest recovery in these sites. We also found that the timing of tree planting matters, but that the importance of timing depends on the level of competition from shrubs. In places where shrub competition is intense, tree planting is much more successful if planting occurs the year immediately following a fire, the soonest that it is usually practical to plant. In contrast, in places where shrub competition is weaker, delaying tree planting until some shrubs establish can facilitate tree seedling survival, perhaps because shrubs provide shelter from harsh conditions. We also found that tree planting was strongly associated with a higher proportion of pine seedlings, although this positive planting effect was weaker when planting happened later after fire, and was cancelled out at higher levels of shrub cover and shrub height.

Rising atmospheric CO(2 )levels place terrestrial ecosystems under novel environmental conditions, and research in field settings is key to understanding how real plant communities will respond. Despite decades of progress in elevated CO2 (eCO(2)) experiments, major gaps persist in our knowledge of plant responses to interacting influences of climate change, especially in areas outside North America and Western Europe. With a goal to expand access to field-based eCO(2) research, we designed, built, and tested TinyCO(2), a low-cost field experiment for climate change research on plants. TinyCO(2) features sixteen 0.62-m(2) plot areas, half with ambient and half with elevated (+200 ppm) CO2 concentrations, and is suitable for short-stature plants (<= 0.5 m in height). Using a proportional-integral control algorithm and constant sampling of air within the plots, TinyCO(2) achieves consistent elevation of [CO2] averaging +196.9 ppm. During testing, 95.1% of measured CO2 concentrations fell within 20% of the setpoint (ambient CO2 + 200 ppm). A streamlined design and efficient use of instrumentation reduced the cost of the system to roughly one-fifth of the cost of similar experiments from the past 30 years ($13.68 vs. $64.65 ppm-1 m(-2), adjusted to 2024 USD). Our results demonstrate a system capable of precise and accurate field-based CO2 elevation for significantly reduced cost. We envision the TinyCO(2) design being implemented in a multitude of field-based eCO(2) studies, perhaps as part of a globally distributed collaborative network experiment.

In the western US, restoring forests to historical, fire-adapted conditions can reduce fire risk, but most fireadapted restoration targets in California focus on the western slope of the Sierra Nevada mountain range. By comparison, Eastern Sierra Jeffrey pine forests experience distinct climate and growing conditions, and both their historical structure and the effects of fire suppression are different from those of their west-side counterparts. In this study, our goals were (1) to use spatially-explicit forest reconstruction methods to estimate the historical, fire-adapted structure of eastern Sierra Jeffrey pine forests; (2) to describe the structural changes observed after 54-65 years of fire exclusion; and (3) to quantify the structural effects and restoration potential of 1-2 recent fire events that followed the fire exclusion period. At two sites in the Eastern Sierra near Mammoth Lakes, CA, we surveyed stands and collected tree ring series to establish tree ages and reconstruct maps of the forest at the end of the frequent-fire period (1941), after the period of fire exclusion (1995/2006), and in recent years (2018). We found that half a century of fire exclusion led to denser stands with fewer, smaller openings and larger clumps of trees, much like in the western Sierra. However, east-side forests were not as departed from fireadapted conditions, and 1-2 fire events in recent years showed potential to restore many structural characteristics to their prior state. Our results can inform forest management decisions, and they support the use of prescribed fire and managed wildfire in forest restoration.

Rising global fire activity is increasing the prevalence of repeated short-interval burning (reburning) in forests worldwide. In forests that historically experienced frequent-fire regimes, high-severity fire exacerbates the severity of subsequent fires by increasing prevalence of shrubs and/or by creating drier understory conditions. Low- to moderate-severity fire, in contrast, can moderate future fire behavior by reducing fuel loads. The extent to which previous fires moderate future fire severity will powerfully affect fire-prone forest ecosystem trajectories over the next century. Further, knowing where and when a wildfire may act as a landscape-scale fuel treatment can help direct pre- and post-fire management efforts. We leverage satellite imagery and fire progression mapping to model reburn dynamics within forests that initially burned at low/moderate severity in 726 unique fire pair events over a 36-year period across four large fire-prone Western US ecoregions. We ask (1) how strong are the moderating effects of low- to moderate-severity fire on future fire severity, (2) how long do moderating effects last, and (3) how does the time between fires (a proxy for fuel accumulation) interact with initial fire severity, day-of-burning weather conditions, and climate to influence reburn severity. Short-interval reburns primarily occurred in dry- and moist-mixed conifer forests with historically frequent-fire regimes. Previous fire moderated reburn severity in all ecoregions with the strongest effects occurring in the California Coast and Western Mountains and the average duration of moderating effects ranging from 13 years in the Western Mountains to >36 years in the California Coast. The strength and duration of moderating effects depended on climate and initial fire severity in some regions, reflecting differences in post-fire fuel accumulation. In the California Coast, moderating effects lasted longer in cooler and wetter forests. In the Western Mountains, moderating effects were stronger and longer lasting in forests that initially burned at higher severity. Moderating effects were largely robust to fire weather, suggesting that previous fire can mediate future fire severity even under extreme conditions. Our findings demonstrate that low- to moderate-severity fire buffers future fire severity in historically frequent-fire forests, underlining the importance of wildfire as a restoration tool for adapting to global change.

Current workflows in academic ecology rarely allow an engagement of ecologists with philosophers, or with contemporary philosophical work. We argue that this is a missed opportunity for enriching ecological reasoning and practice, because many questions in ecology overlap with philosophical questions and with current topics in contemporary philosophy of science. One obstacle to a closer connection and collaboration between the fields is the limited awareness of scientists, including ecologists, of current philosophical questions, developments and ideas. In this article, we aim to overcome this obstacle and trigger more collaborations between ecologists and philosophers. First, we provide an overview of philosophical research relevant to ecologists. Second, we use examples to demonstrate that many ecological questions have a philosophical dimension and point to related philosophical work. We elaborate on one example – the debate around the appropriate level of complexity of ecological models – to show in more detail how philosophy can enrich ecology. Finally, we provide suggestions for how to initiate collaborative projects involving both ecologists and philosophers.

High severity fire is far outpacing post-fire reforestation capacity across the American West, highlighting the need to better understand when and where natural regeneration is sufficient to meet short- and long-term reforestation goals. The vast majority of available post-fire data represents regeneration in the first few years following a fire, raising the question of how well recovery trajectories can be predicted from early post-fire snapshots. We utilize a unique dataset from seven wildfires and 78 plots surveyed twice following stand- replacing fire in two California ecoregions, the northern Sierra Nevada and Klamath Mountains, to ask (1) how are post-fire vegetation and conifer densities changing over time and (2) how well do relatively early post- fire sampling efforts and biophysical factors (availability of seed source, shrub competition, and microclimate) explain longer-term (12-23 year) conifer recovery in California dry forests. Change in conifer seedling density between survey periods was highly variable. Overall, densities of all conifer species during the first decade after fire provided reliable estimates of longer-term conifer recovery in northern California dry forests, though in different ways for different species. Early post-fire seedling density readily explained longer-term densities for less shade-tolerant species, yellow pine ( Pinus ponderosa and Pinus jeffreyi) ) and Douglas-fir ( Pseudotsuga menziesii). ). In contrast, early seedling density was a poor predictor for shade-tolerant species, predominantly white fir ( Abies concolor), ), with site conditions better explaining longer-term density. In general, the probability of a site experiencing net seedling mortality between surveys was highest in sites with high shrub cover and the probability of net recruitment increased with decreasing heat load, suggesting a continued role of microclimate and competition in forest recovery long after the initial post-fire period. Our results lend support to the use of relatively early post-fire surveys to infer longer-term forest recovery trajectories for forest management planning and can help refine reforestation prioritization tools.

Despite the objective of early 20th century foresters to regenerate light-loving pines in mixed-conifer forests of the western United States, pine co-dominance has declined over the last century. Here, we investigate the decline of pine by combining historical and contemporary tree regeneration data from a repeat-measured Sierra Nevada mixed-conifer forest to evaluate species-specific post-logging establishment success and track subsequent stand outcomes over more than a century of fire exclusion. We leveraged this permanent plot network to quantify 1) changes in regeneration abundances among species over an 88-year period, 2) the influence of microsite conditions on seedling abundances and 3) whether post-logging seedling data can predict future forest composition. We found that the relative abundances of both pine and white fir regeneration decreased over time (from 12% to 5% and 60% to 46% respectively) while incense-cedar abundance increased (from 32% to 49%) far beyond its representation in the overstory (17% of basal area). Despite pine’s known positive response to high light availability, we did not find strong evidence for the positive influence of gaps on pine regeneration in any time period, even following creation of gaps with logging. However, across species, post-logging skid trails were positively associated with regeneration and woody debris was negatively associated with regeneration in at least one time period. Finally, we discovered that the occurrence of pre-logging advance regeneration best predicted new contemporary trees across all species. For white fir and incense-cedar, regeneration from up to ten years post-logging was also associated with new contemporary trees. For pine, no seedling that established in the post-logging period transitioned into the contemporary canopy. Notably, however, even the pre-logging regeneration layer contained very little pine (12%), driving a mismatch with nearby overstory pine which comprised 32% of tree basal area. Such a mismatch is likely an early legacy of fire exclusion: by 1928, substantial increases in shade tolerant species had shifted the composition of the regeneration layer and thereby the future forest. Overall, our work contributes to growing evidence that increasing light availability alone is not enough to promote pine regeneration. Other ecosystem conditions associated with the historical fire regime—including availability of disturbed substrate, lower density of woody debris, the presence of large-diameter pines as seed sources, and selection against shade-tolerant species by predominantly low severity fire—are essential for promoting pine co-dominance in mixed-conifer systems.

Wildfires may facilitate climate tracking of forest species moving upslope or north in latitude. For subalpine tree species, for which higher elevation habitat is limited, accelerated replacement by lower elevation montane tree species following fire may hasten extinction risk. We used a dataset of postfire tree regeneration spanning a broad geographic range to ask whether the fire facilitated upslope movement of montane tree species at the montane-to-subalpine ecotone. We sampled tree seedling occurrence in 248 plots across a fire severity gradient (unburned to >90% basal area mortality) and spanning ~500 km of latitude in Mediterranean-type subalpine forest in California, USA. We used logistic regression to quantify differences in postfire regeneration between resident subalpine species and the seedling-only range (interpreted as climate-induced range extension) of montane species. We tested our assumption of increasing climatic suitability for montane species in subalpine forest using the predicted difference in habitat suitability at study plots between 1990 and 2030. We found that postfire regeneration of resident subalpine species was uncorrelated or weakly positively correlated with fire severity. Regeneration of montane species, however, was roughly four times greater in unburned relative to burned subalpine forest. Although our overall results contrast with theoretical predictions of disturbance-facilitated range shifts, we found opposing postfire regeneration responses for montane species with distinct regeneration niches. Recruitment of shade-tolerant red fir declined with fire severity and recruitment of shade-intolerant Jeffrey pine increased with fire severity. Predicted climatic suitability increased by 5% for red fir and 34% for Jeffrey pine. Differing postfire responses in newly climatically available habitats indicate that wildfire disturbance may only facilitate range extensions for species whose preferred regeneration conditions align with increased light and/or other postfire landscape characteristics.

Predicted increases in extreme droughts will likely cause major shifts in carbon sequestration and forest composition. Although growth declines during drought are widely documented, an increasing number of studies have reported both positive and negative responses to the same drought. These divergent growth patterns may reflect thresholds (i.e., nonlinear responses) promoted by changes in the dominant climatic constraints on tree growth. Here we tested whether stemwood growth exhibited linear or nonlinear responses to temperature and precipitation and whether stemwood growth thresholds co-occurred with multiple thresholds in source and sink processes that limit tree growth. We extracted 772 tree cores, 1398 needle length records, and 1075 stable isotope samples from 27 sites across whitebark pine's (Pinus albicaulis Engelm.) climatic niche in the Sierra Nevada. Our results indicated that a temperature threshold in stemwood growth occurred at 8.4°C (7.12-9.51°C; estimated using fall-spring maximum temperature). This threshold was significantly correlated with thresholds in foliar growth, as well as carbon (δ13 C) and nitrogen (δ15 N) stable isotope ratios, that emerged during drought. These co-occurring thresholds reflected the transition between energy- and water-limited tree growth (i.e., the E-W limitation threshold). This transition likely mediated carbon and nutrient cycling, as well as important differences in growth-defense trade-offs and drought adaptations. Furthermore, whitebark pine growing in energy-limited regions may continue to experience elevated growth in response to climate change. The positive effect of warming, however, may be offset by growth declines in water-limited regions, threatening the long-term sustainability of the recently listed whitebark pine species in the Sierra Nevada.

Many dry conifer forests in the western United States were historically adapted to frequent low-to-moderate severity fires, but are increasingly susceptible to large, stand-replacing wildfires due to dramatically altered stand conditions and changing climate. The historic tree spatial patterns of mature stands in fire-adapted forests - individual trees, clumps of trees, and openings (ICO) - are associated with heightened resistance and resilience to fire. How this pattern develops over time, however, is not well understood and could help inform reforestation practices better designed to increase fire resistance in developing stands. We investigated growth rates and spatial patterns among regenerating trees in mixed-conifer forests with restored fire regimes in California's Sierra Nevada. We compared average stocking densities across tree species, size classes, shrub cover, and fire histories. We also examined the effects of microsite topography on spatial patterning of these juvenile trees, and the effects of clump patterning, local stem density and adjacent shrubs on tree growth rates. We found that the majority (75%) of sampled stems were found in clumps. Our mixed-effect models indicated that for trees growing within clumps, increased crowding slowed tree growth, as expected. Surprisingly, however, compared with individual trees growing outside clumps, trees growing within clumps grew significantly faster. Shrub cover in proximity to juvenile trees did not have a consistent impact across our models, but was associated with increased annual height growth. Additionally, plots with high shrub cover had higher stocking rates among the tallest regenerating stems (height > 137 cm). Our findings indicate that clumped spatial patterns of natural tree recruitment may favor the establishment and early growth of regenerating conifers in active-fire forests. While our study focused only on the early stages (<30 years old) of regeneration, our results contrast with common reforestation strategies favoring regular, widely-spaced plantings and aggressive shrub reduction. Our research suggests we need a better understanding of how heterogeneity in the spatial patterns of juvenile trees and shrubs may enhance the resilience of regenerating stands as they mature.